A study of cooperative learning group dynamics in an inquiry-based physics course
by Meredith Jennings Wills-Davey
A thesis submitted in partial fulfillment of the requirements for the degree of Master of Science Physics
Montana State University
© Copyright by Meredith Jennings Wills-Davey (1999)
Abstract:
We consider the nature of group dynamics in small cooperative learning groups in the inquiry-based
physics course Physics 201: Physics by Inquiry. Our study was motivated by observed changes in the
behavior of some students as they participated in various groups during the Spring 1998 semester. In
order to understand the origin of these behavior changes, approximately half of the class was
interviewed concerning their group experiences.
The results of these interviews suggested that students could be categorized into five different group
learning styles, and that certain combinations of learning styles led to positive’ productive group
dynamics, while other combinations led to internal group friction. The strongest predictor of group
interaction appeared to stem from a student’s previous experience with the course subject matter;
students with some background in. physics had extreme difficulty working with novice learners. A less
strong but noticeable predictor related to the students’ self-confidence as learners. Students without
previous physics knowledge could be classified into four levels of confidence, and it was found that
extremely confident students had difficult working with extremely insecure students.
We formed hypotheses for organizing cooperative learning groups with successful group dynamics
based on this categorization scheme, and applied them to two sections of Physics by Inquiry in the Fall
1998 semester. Students from this semester were then also interviewed. We found that, when our
grouping strategies were implemented, group . success rates increased from.36% in the Spring 1998
semester to 75% in the Fall 1998 semester.
However, data from the Fall 1998 semester revealed several secondary problems. Non-traditional
students were observed to conflict with traditional students in certain circumstances. When students
with previous physics experience were grouped together, problems with the curriculum became
apparent. Finally, assigning multiple groups throughout the semester was found to have negative
effects on student interactions. Our original hypotheses were modified to help rectify these problems.
We conclude that our cooperative grouping strategies are extremely effective for creating groups with
positive and productive dynamics. We compare them to grouping techniques used in other studies, and
we discuss the potential for wider implementation. A STUDY OF COOPERATIVE LEARNING GROUP DYNAMICS
IN AN INQUIRY-BASED PHYSICS COURSE
by
Meredith Jennings Wills-Davey
A thesis submitted in partial fulfillment
o f the requirements for the degree
of
Master o f Science
in
Physics
MONTANA STATE UNIVERSITY-BOZEMAN
Bozeman, Montana
November 1999
ii
APPROVAL
o f a thesis submitted by
Meredith Jennings Wills-Davey
This thesis has been read by each member o f the thesis committee and has been
found to be satisfactory regarding content, English usage, format, citations, bibliographic
style, and consistency, and is ready for submission for the College o f Graduate Studies.
Dr. Jeffrey Adams
x /C
?
^
(Date)
Approved for the Department of Physics
K-T-M
Dr John Hermanson
(^Signature)
(Date)
Approved for the College o f Graduate Studies
D
Drr Bruce R McLeod
(Signature)
(Date)
iii
STATEMENT OF PERMISSION TO USE
In presenting this thesis in partial fulfillment o f the requirements for a master’s
degree at M ontana State University-Bozeman, I agree that the Library shall make it
available to borrowers under the rules o f the Library.
If I have indicated my intention to copyright this thesis by including a copyright
notice page, copying is allowable only for scholarly purposes, consistent with “fair use” as
prescribed in the U S . Copyright Law. Requests for permission for extended quotation
from or reproduction o f this thesis in whole or in parts may be granted only by the
copyright holder.
Signature
D ate U - S jgI cI
iv
ACKNOWLEDGEMENTS
This work would not have been possible without contributions from a variety o f
sources. I would like to thank the students who agreed to be interviewed, both for their
candor and the valuable information they provided:
BriAnne Burgess
Kari Chaffin
Reid Christensen
Juline Cox
Laura Depaoli
Brielle Erickson
Stephanie Ewald
Malinda Gates
Sheila Gobert
Richelle G off
Christi Hargis
Robyn Harris
Renae Hofer
Kelley Hurley
Naomi Kelm
Anne Kent
Melissa Kuhs
Alethea Lambert
Jacqueline Larson
Nancy Lenhardt
Molly McLain
Buck Measure
TaraM oeller
Ana Morris
Michelle Nazelrod
Darci Owen
Kathleen Picard
Jennifer Posivio
M argaret Richardson
Alan Roos
Michelle Shadel
David Shobe
Deborah Solberg
Katharine Taylor
John Thomley
Lindsey Turner
Rachel Turner
Breanna Underwood
Erin Washburn
Holly Wheelock
I would also like to thank the graduate students, staff, and faculty who donated
their time as interviewers:
Jeffrey Adams
Patrick Bemathy
Christine Brick
Thomas Brown
Rebecca Lindell Adrian
Gregory Francis
Timothy Slater
George Tuthill
Melissa Wright
Finally, I would like to thank Jefferson Crowder, who served as my co-instructor during
the Fall 1998 semester o f Physics by Inquiry, for allowing me to implement my hypotheses
on cooperative group dynamics in his classroom.
TABLE OF CONTENTS
Page
1. IN T R O D U C T IO N ............................................................................................. ............
I
A B rief Description o f Physics Education R esearch..............................................
The Motivation for our S tu d y .......................... ...... .............................................
I
3
2. EXAM INING STUDENT B E H A V IO R ............ .........................................................
5
P ro c e d u re ...................................................................................................
D a ta ............................................................................................... :..............................
Personality D escriptions....................
Hierarchy Descriptions .......................................................
Types o f Students Observed ................ ....................................... ......................
Previously-knowledgeable.............................
Aggressive-confident ...........................................
Cooperative-confident ................................................................................
Cooperative-uncertain .............................................................................i'..
Passive-uncertain ..............................................
Problems Uncovered by the Interviews ............................................. ................. .
Group Pace .............................................................................................................
Extreme Heterogeneity .............
Group Success R a te .............................................................
5
6
6
8
11
11
14
16
18
19
21
21
23
28
3. OUR RESULTS VERSUS RESULTS OF PREVIOUS S T U D IE S .......... .............. 31
Reviewing Comparable Physics Education Group-work Research ..................... 33
Towards the Implementation o f Successful Cooperative Groups
. in Inquiry-based L earning......................................................
43
4. IM PLEM ENTATION AND FO LLO W -U P................................................................ 46
Results o f the Follow-up Study .....................................
Group Success Rates ..........................................................
49
49
vi
TABLE OF CONTENTS— Continued
Page
Demonstrations o f Improved Group Dynamics ......................................................
Extreme H eterogeneity.................................... ;.................................................
Diversity in Group P a c e ........................................................................................
Presence o f a Distinct L e a d e r................................................... ........................
Problems Arising during Implementation.............................................. ............. ..
Non-traditional Students......................... ........................................'..................
Previously-knowledgeable Students and Expected Coursework ...................
Group-switching Backlash ............................................................;.....................
Improvements for Future Im plem entation...............................................................
5. D IS C U S S IO N .........................................
54
54
57
60
62
62
68
71
76
79
Conclusions o f our S tu d y .......................................................................................
79
Implications for Use in Other Environments
............................................ ...... 83
Future Study ................................................................................................................ 86
B IB LIO G R A PH Y ...............................................................
88
APPENDIX: DISCUSSION OF THE STUDENT
CATEGORIZATION IN STR U M EN T..................................................
92
Instrument D evelopm ent..............................................................
Problems with the Categorization Instrum ent....................................
93
96
vii
LIST OF TABLES
Table
1. Group Composition in Relation to Group Success
for the Spring 1998 Semester...........................................................
Page
29
2. Distribution o f Students According to Cooperative
Group Learning Categorization for the Fall 1998
Semester......-..............;..............................................................................................47
3. Organization o f Cooperative Learning Groups
according to Group Learning Classification for
the Fall 1998 Semester................................
48
4. Group Success Indexes for the Fall 1998 Semester............................... ................53
viii
LIST OF FIGURES
Figure
Page
1. Student Group Learning Style Categorization
Question Tree............................................................................................................. 94
2. Instrument Implemented to Determine Group ,
Learning Categorization for the Fall 1998
Sem ester..................................................................................................................... 95
3. Modified Student Group Learning Style
Categorization Question Tree..................................................................................99
ix
ABSTRACT
We consider the nature o f group dynamics in small cooperative learning groups in
the inquiry-based physics course Physics 201: Physics by Inquiry. Our study was
motivated by observed changes in the behavior o f some students as they participated in
various groups during the Spring 1998 semester. In order to understand the origin o f
these behavior changes, approximately half o f the class was interviewed concerning their
group experiences.
The results o f these interviews suggested that students could be categorized into
five different group learning styles, and that certain combinations o f learning styles led to
positive’ productive group dynamics, while other combinations led to internal group
friction: The strongest predictor o f group interaction appeared to stem from a student’s
previous experience with the course subject matter; students with some background i n .
physics had extreme difficulty working with novice learners. A less strong but noticeable
predictor related to the students’ self-confidence as learners. Students without previous
physics knowledge could be classified into four levels o f confidence, and it was found that
extremely confident students had difficult working with extremely insecure students.
We formed hypotheses for organizing cooperative learning groups with successful group
dynamics based on this categorization scheme, and applied them to tw o sections o f
Physics by Inquiry in the Fall 1998 semester. Students from this semester were then also
interviewed. We found that, when our grouping strategies were implemented, group .
success rates increased from.36% in the Spring 1998 semester to 75% in the Fall 1998
semester.
However, data from the Fall 1998 semester revealed several secondary problems.
Non-traditional students were observed to conflict with traditional students in certain
circumstances. When students with previous physics experience were grouped together,
problems with the curriculum became apparent. Finally, assigning multiple groups
throughout the semester was found to have negative effects on student interactions. Our
original hypotheses were modified to help rectify these problems.
We conclude that our cooperative grouping strategies are extremely effective for
creating groups with positive and productive dynamics. We compare them to grouping
techniques used in other studies, and we discuss the potential for wider implementation.
I
J
CHAPTER I
INTRODUCTION
A B rief Description o f Physics Education Research
Physics has a reputation as a difficult academic subject, both to Ieam and to teach.
As one o f its main goals, the burgeoning field o f physics education research strives to
make physics more accessible to students. This is being done largely through a
reexamination o f the nature o f subject material, the effectiveness o f teaching techniques,
and the appropriateness o f assessment procedures. Progressive instructors are moving
tow ards more conceptual - rather than formula-based mathematical - understandings o f
physics concepts. The more didactic methods o f instruction, such as large-enrollment
lectures or recitation sessions, are giving way to classrooms in which students are more
actively involved. Student discussion is encouraged, with students often being challenged
to question and justify results. Laboratory activities are moving away from exercises that
merely verify well-known experiments, and instead are focusing on creating cognitive
conflicts which challenge students to confront and resolve their own non-Newtonian
preconceptions.
One teaching environment that has gained popularity in the physics education
community is the small cooperative learning group. These groups typically consist o f 3-4
students, and are organized with the expectation that the students within the groups will
w ork collectively toward some common goal, such as solving problems, producing a
group project, or learning designated subject material. The use o f cooperative learning
groups is not new. There are relevant and compelling sociological studies dating back to
the mid-1950s. However, with the recent onset o f physics education research as a
recognized field within the discipline o f physics, the physics community has renewed
interest in this style, o f learning in the hope o f improving student understanding.
Myriad studies have examined the effectiveness o f cooperative learning groups for
teaching physics. Since laboratory work already has a history o f using small groups, many
o f the recent studies (Cottle & Lunford, 1995; Emkey, 1979; Heller & Hollabaugh, 1992;
Heller, Keith, & Anderson, 1992; etc.) have used cooperative learning groups as a
supplement to more traditional lecture-based teaching methods, using the discussion
format to focus on problem-solving skills. In most o f these cases, students were organized
to w ork quantitative problems individually with help from fellow group members. In some
other cases, the group was assigned to work collaboratively on a single solution to a
posed problem.
Other research (Shaffer & McDermott, 1992, for example) examined cooperative
groups as a potential format for all instruction, with a strong focus on conceptual
understanding. Here, rather than the more free-form discussion o f problem-solving
sessions, instructors would interact with the students through Socratic dialogue, providing
an environment o f guided inquiry. At least two curricula have been created that take
students through the guided inquiry process: M cDermott’s Physics by Inquiry and Law s’
W orkshop Physics.
The Motivation for our Study
This study was inspired by an instructor’s observation o f student behavior during
the Spring 1998 semester in a non-lecture-based physics course using cooperative group
learning and guided-inquiry techniques to teach physics. This particular physics class was
Physics 201: Physics by Inquiry, a course offered at Montana State University - Bozeman
primarily for pre-service elementary school teachers using curricula derived from
M cDermott’s Phvsics bv Inquiry. The population o f the section under study was
predominantly female - seventeen women and two men - and included many students who
were extremely unfamiliar with, and in some cases even “phobic” of, physics. Students
were for the most part o f traditional age, and with the exception o f one Native American
student, all were Caucasian.*
While cooperative group learning had been used since the inception o f the course
in 1995, effective techniques o f student grouping were not considered before this study.
Students created self-formed groups on the first or second day o f class* and remained in
those groups throughout the semester. In the Spring 1998 semester, the instructors
implemented mandatory group switching to allow students to interact with members o f the
class outside their initial groups. The initial groups were still self-formed, but the three
* Because o f the lack of gender and ethnic diversity, we did not. consider these variables separately in our
study.
t For some semesters, the first day o f class is devoted to social activities, so that students can have some
interaction with all o f the other members of the class before choosing groups.
4
subsequent rounds o f groups were instructor-assigned. There was no particular instructor
motivation behind grouping assignments.
W hen the instructors reassigned groups, they noticed obvious changes both in
individual student behavior and in overall group dynamics. Some students appeared more
dominant or passive depending on their group, composition. Others seemed engaged and
interactive in one group, only to instigate conflict in another. To understand these
observed “personality” changes, nine students were interviewed from the Spring 1998
semester. Surprisingly, the information provided in these interviews proved more useful
as a diagnostic o f group dynamics than as a window into the temperaments o f individuals.
Unmistakable patterns emerged, which suggested that the nature and success o f dynamics
in small cooperative learning groups was not only understandable, but often predictable..
The aforementioned informal observations motivated the principle research
question for this study: Based on limited initial student-provided data, is it possible to
organize cooperative learning groups in which all students will be engaged, productive
contributors and learners? The following chapters present the results o f this study.
Chapter 2 offers data and conclusions derived from the interviews o f the Spring 1998
students. Chapter 3 compares our results with those o f other studies o f cooperative
learning groups, and discusses similarities and differences. For Chapter 4, we implement
the hypotheses derived in Chapter 2 in two sections o f the Fall 1998 offering o f Physics by
Inquiry and discuss our findings. Chapter 5 offers our conclusions and implications for
use o f our findings by other instructors.
'5
CHAPTER 2
EXAMINING STUDENT BEHAVIOR
Procedure
Our study was motivated by observations o f changes in student behavior as groups
were reorganized. Nine undergraduate students - out o f a class o f nineteen - were
interviewed about their group learning experiences in Physics 2 0 1. Six o f the interviewees
w ere selected because the instructors noted their behavior changed from group to group.
The other three students appeared to behave consistently in all groups, and were chosen to
function as a quasi-control group. The interviewers had no contact with the students
before or after their interviews. The interviews were performed individually with the
interviewers working from a script. Students were prompted to describe their personal
experiences in their first three learning groups. They were asked specifically about how
they interacted with other students in their groups. The interviewer focused the questions
on the various roles each student played, their subsequent behaviors, and on the hierarchy
that formed within each group. Students were also asked to describe positive or negative
experiences within each group.
The interviews were audio-taped and transcribed. The investigator organized
quotes from the interviews into three overarching categories: participation, hierarchy, and
group dynamics. These categories often overlapped and were used mainly to group
6
comments about similar topics. The “participation” category focused on the level o f
involvement o f each student in a group; the “hierarchy” category discussed issues such as
leadership and general group organization; and the “group dynamics” category regarded
the social interactions between group members.
Because the investigator was interested in changes in students’ roles from group to
group, students’ personality traits were cross-referenced. If tw o interviewees provided
radically different descriptions o f a particular student’s behavior in different group
settings, the motivation for these personality changes would be considered; if not, it would
suggest that group organization alone led to changes in a student’s behavior.
Data
Personality Descriptions
In most cases, student descriptions o f fellow group members’ behavior were
consistent, regardless o f group or personal relationship. Negative traits received more
attention than positive ones, and most interviewees had the same negative reactions to the
same students. Thus, there were considerably more data on some students than others. In
such detailed cases, no external descriptions proved incongruous, and the only
inconsistencies in self-reports arose where students were not aware o f their own negative
characteristics. As an example, one student is described by members o f her three different
groups:
7
The girl that liked to ask the teacher the answers to questions, she liked to stop us
a lot, ju st fo r the fa c t that she didn 't really like to think fo r herself, and so w e 'd
say, “Well, think about it. This is... ” and she ’d say, well, no, she wanted to ask
the question.
Member o f student’s first group
[W]e would be trying to explain [a concept] to her... PjJut it seemed like she
■always had to return to phe instructor], instead o f asking, instead o f a group
discussion o f what we all thought.
Member o f student’s second group
[I]f there wasn't something that she understood, she wouldn 't confide in us so
much, but would ask the TA [ teaching assistant] more, rather than trying to work
it out within the group or something.
Member o f student’s second group
S h e'd ask a question, and w e’d set out to try and explain it, and we 'd get the first
couple sentences out and sh e’d like, “[Instructor's name] ! ”.... A nd that kind o f
irritated me a little bit because she wouldn’t let us get our ideas out.
Member o f student’s second group
[ I ] f it ’s [the interviewee 7S answer] not what she got, then she doubts herself. You
know, she 7Il be like, “That can 7t be right, ’’-...and then usually calls fo r a third
opinion.
Member o f student’s third group
Even her self-description was consistent, albeit somewhat more charitable:
I think the only way I ran into problems was when they fe lt like they knew the
answer, like, they knew that they had the right answer, and I doubted. Iw as kind
of, like, edgy, and so other people in the group were kind o f edgy, s o l ’d usually
ask, and I think that would frustrate some o f them when they were right. They 7re
like,. “Just listen to me. !have the answer.77 So that was kind o f frustrating to
them, that sometimes it would help to ask the teacher, ‘cause they 7d p u t it into
different terms.
Self-description
W hen student descriptions diverged, it stemmed from a lack o f detailed description
on the part o f some interviewees. In particular, this seemed to occur when a student had
8
failed to leave an impression. In one instance, some interviewees offer personal
descriptions o f a student:
Our other group member kind o f played the passive role, but she thought o f
questions that we never asked, really deep thought questions maybe? She would
come up with a question out o f the blue sometimes.
Member o f student’s second group
AS? kind o f works at a pace between the two o f us. She ’ll be with BA as fa r as
moving at the same pace, but then she and I work similarly in how we get our
answers. We think real similarly.
Member o f student’s third group
while others discuss her only in vague, generalized terms:
Mostly, they [the other group members, which included this student] did all the
questions.
Member o f student’s first group
[EJverybqdy else [the other group members, including this student] kind o f
watched, mostly because they liked to chat.
Member o f student’s second group
Hierarchy Descriptions
Interviewees were asked to describe themselves and fellow group members in the
context o f group roles and, specifically, a group hierarchy. Two repeating themes
emerged. In most groups, a specific hierarchy was present and students described it as
static:
INTERVIEWER: Do the roles change very often, or do you guys all kind o f do
the same thing all the time?
STUDENT: !w ould say, well, in the three groups I ’ve had, they probably d o n ’t
change much.
* When students refer to each other by name in interviews, a two-letter anonymous representation o f the
student’s name has been used. Two-letter representations were assigned randomly to each student, such
that they have no relation to the student’s actual name, and are used self-consistently throughout this
paper.
9
In these cases, all the interviewees in a given group identified the same student(s) as the
leader(s). Although the actions o f the leader varied from group to group, he or she was
considered the “smartest,” or most knowledgeable in the group:
ND, I guess, would be kind o f the leader in that group, because she was more
boisterous in her ideas, and the rest o f us were kind of, like, holding back,
because none o f us had had any o f that in dur school.
I would say that R S was probably the. leader. She would go out and get .
everything andjust start putting things together. It ju st clickedfor her.
It is interesting to note that the person students named as the leader didn’t necessarily
contribute positively to the group, and in some cases, serious problems emerged:
STUDENT: They [ the leaders] were so fast-paced, and I kind o f got it. They ju st
kept working and I fe lt like I had to work extra-fast to keep up...
INTERVIEWER: D oyou think they knew this?
STUDENT: Yeah. It didn’t matter.
[Y jo u 'd say one [an opinion], but she 'd never think that yours was right, even i f
yours actually was, so it was hard to convince her. The other three got along a
lot better, because we would, like, talk about it, whereas she 'd ju st like, “No, i t ’s
this way. ” You ’re like, “No, really, it isn V. ”
HC, in this group, went ahead o f everyone. I guess he ’d look at the question and
say, “Well, that’s easy! ” and write down two things, and raise his hand and ask
[the instructor] or a TA [teaching assistant] to check us off. A n d he was getting
us into trouble fo r a couple classes ‘cause they would say, “Well, your group's
not ready.”
There was a definite leader who always was leading everything andjust took
control, didn’t really give anyone else a chance to do that.
ND would already know it. S he'dju st write down an answer, ju st right away, but
to me it fe lt like she knew everything and her answer was “the word, ” and so
she ’d write it down. That’s where she would get into trouble a lot; she wouldn 't
discuss it and stuff, and [the instructor] would come over and say, “Well, how did
you get this answer? ” and it was not the right way o f doing it.
In fact, students more often reported negative experiences with group leaders than
positive ones.
Beyond group leader, other student roles were more difficult to classify. Some
students described themselves as skeptics or questioners and some acknowledged their
limited contributions to group discussion:
I was always considered the quiet one.
[SJometimes I 'd ju st kind o f sit there, and then they 'd notice that I didn’t know
what was going on, and they'd help me. '
On the whole, however, no group roles consistently stood out other than the group leader.
In tw o o f the groups, a distinct lack o f leadership was present. In these cases,
students failed to fall into definite roles and no hierarchy emerged:
STUDENT: A t any given time, we all contributed pretty well. I think, i f one
person started doing too much, the others would take part more. A nd that
happened, I think. One person maybe setting it all up at one time, but then
another would try to get involved more.
INTERVIEW ER: A nd it could have been anybody? It wasn't sort o f one person
kind o f running the show?
STUDENT: No, no. Right, right.
INTERVIEW ER: Soyou took on more o f a leadership role in the second group?
STUDENT: It was definitely more shared.
INTERVIEW ER: So, is there any hierarchy you would give to that second group?
STUDENT: No, not in that group.
In groups that lacked hierarchy, students instead described group roles as moving naturally
between members:
I thought it [the leadership hierarchy] changed quite a bit. Sometimes people
would initiate different activities. I ’d come in, one person would get it going one
day and we would start following along, and there might be another person the
next day.
11
We all took turns depending on what section we were working on. In one section,
M W w ouldhavebeenm oreofaleader...
It is important to note that students consistently described these non-specific role groups
as their favorites.
Types o f Students Observed
In interviewees’ descriptions o f themselves and others, clear learning and
participation styles1" emerged. Even if students’ own roles changed from group to group,
their basic attitudes towards learning and their interactions with other people did not vary
significantly Based on descriptions from our nine interviewees, we found that the
nineteen students in the course displayed five distinct sets o f traits. W e named them as:
previpusly-knowledgeable, aggressive-confident, cooperative-confident, cooperativeuncertain, and passive-uncertain. These categories were derived more from group
personality than academic prowess. In fact, we found no relation between academic
achievement and group confidence.1
Previouslv-knowledgeable Two interviewees and one non-interviewee stood out
due to their previous knowledge o f and comfort level with physics. In this case, all three
had had positive experiences in high school physics courses. The tw o interviewees saw
t Our use o f the term “learning styles” refers to the way in which students seem to Ieam within the
cooperative learning group environment. We do not consider it related to similar terminology - such as
the “cognitive styles” used by Miller, Wilkes, & Cheetham (1993) - althpugh possible correlations
between our observed “group learning styles” and other typse o f learning styles may exist,
t A natural bias seems to exist, especially in group learning, that people who are outgoing and selfconfident are naturally higher achievers; a lack o f talking conveys a lack o f understanding. In fact, the
instructors found that it was not unusual for a particularly withdrawn student to be academically
successful (perhaps due to oyerzealous studying stemming from insecurity with the subject matter) or an
outgoing student to perform poorly (perhaps due to overconfidence with the material).
12
themselves as more knowledgeable and, in some cases, smarter than their fellow group
members:
With my first group, it was currents and electric circuits and stuff, and I
understood that... I know how some things shouldfunction and how they should
not, ju st from that knowledge, and then my other group members pretty much
didn ’t have a clue.
Previously-knowledgeable student
I think the reason, mostly, that I assume the leadership role is that I have the most
science background.
Previously-knowledgeable student
[Mfaybe I'm ju st a little bit quicker on the draw...
Previously-knowledgeable student
I ’ve basically done balances and stuff like that since elementary school, so i t ’s
ingrained in me. I know what’s going on; I know what’s supposed to. happen and
there’s a couple members o f my group that I ’m not sure have had that much
experience...
Previously-knowledgeable student
These comments suggest that “previously-knowledgeable students” encountered no new
material in the course at all. One student went so far as to make the statement:
Usually, not everyone would know the answer to the question when it was asked...
Previously-knowledgeable student
*
This statement is interpreted to mean that the activity questions were there to probe the
knowledge o f other group members and not his own.
Other students indicated that they were aware o f the knowledge level o f their
previously-knowledgeable group members and more credibility was given to their
opinions. Each o f these students was the leader in every group he or she participated in
and none o f them were involved in groups that lacked a hierarchy5.
5 One previously-knowledgeable student described her third group as lacking a hierarchy by saying:
13
Previously-knowledgeable students were also characterized by the fast pace at
which they completed tasks. This pace was often resented by other members in their
groups:
<t
They both have so much physics background... That was my hardest group,
probably...
Aggressive-confident student
There was one that, like, really knew it, and he just, like, whipped through it, and
the rest o f us would be like, “H uh?”
Cooperative-uncertain student
N o interviewee described a previously-knowledgeable student as retarding his or her w ork
pace; however, in some cases, previously-knowledgeable students demonstrated patience
with the w ork o f other students and occasionally served as resources.
[He] helped us through some stuff we didn’t understand.
Aggressive-confident student
A nd then sometimes, it was good, because he knew so much, he could explain it a
little better, because you really didn 't get much explanation -w ell, you do when
you ’re trying to check it - but most o f the time i t ’s ju st your own group, and so it
was good to have someone who knew a little more...
Cooperative-uncertain student
With this group, there’s probably an equal balance. There's no one leader. We all take turns
setting up, we all take turns, you know, just “Hey, le t’s do this. ”
Previously-knowledgeable student
However, descriptions o f that same group by another interviewee and in the instructor’s report painted a
very different picture:
ND really took on a dominant role. She fe lt like she knew everything. She would explain
concepts to us; it was like she would talk down to us. And she wouldn ’t give us time to discuss...
Passive-uncertain student
ND was very domineering in dealing with her other groupmates. While a level o f organization
can be productive, ND controlled virtually all o f the work done and conclusions drawn by this
group.... She fe lt the need to run all aspects o f the experiments, sometimes preempting work
already done by her fellow group members.
From instructor’s report
In this case, w e concluded a hierarchy was present despite the student’s self-report.
14
Well, she tried to help us, ‘cause she would say, “What do you understand, FF? ”
and I 'd say, “Yeah, ’’ or else i f I didn’t understand, I 'd say, “Well, I don't
understand how you ’re getting this answer here, ” and sh e'd explain it tome...
Passive-uncertain student
In general, previously-knowledgeable students functioned as group leaders, and tended to
dominate the hierarchy.
Aggressive-confident Two interviewees who possessed no physics background
nonetheless emerged as strong and dominant students. They stood out largely because o f
their self-confidence as learners. Although they had never encountered any o f the physics
concepts, they felt they had a better grasp than other group members and saw themselves
as resources:
[L]ike, in circuits, that was my strongest area, and I fe lt like other people would
look to me.
Aggressive-confident student
I guess I explain things, and i f she has a question. I ’ll step up and try to give her
an example that she ’ll understand or relate it to something else.
Aggressive-confident student
Like the previously-knowledgeable students, they also were comfortable working at a
faster pace than the rest o f the group. They expressed frustration with students who did
not quickly grasp concepts:
fWJhen we start doing stuff, she ’ll say, “Well, I don’t understand. ” I ’m like, “It
says it right there. We ’re going to measure this and measure that. ” I would say
i t ’s a little frustrating.
Aggressive-confident student
15
S h e'd always have to maybe ask a question that’s maybe not needed? I suppose
every time we come up with an answer, or at the end o f an experiment, sh e's like,
“Well, that doesn 't make sense, ’’ or “That can’t be. ” A nd we ’ll say, “Well,
w hat’s wrong? ” or “Why? ” and she kind o f doesn’t know, but she ju st doesn't
think that could be the right answer.
Aggressive-confident student
It appears that most other students failed to differentiate between previously- .
knowledgeable and “aggressive-confident students,” but these students themselves were
keenly aware o f their more experienced classmates. Aggressive-confident students clearly
felt hurried and intimidated by previously-knowledgeable students. M ore than others,
these students perceived the previously-knowledgeable students as impeding their learning
process:
I d o n ’t know i f I fe lt ... / was maybe a little intimidated, but maybe I was thinking
that He is so smart, why should I open my mouth? I might say the wrong thing.
A nd then a fte r -y o u know, like, in the first group, I understood circuits really
well, and then we started going into voltage and all that, and so I understood it
ju st as well, but he kind o f put a cap on everything we had to say, ju st because he
was already two steps ahead o f where we were thinking then.
Aggressive-confident student
Their comfort with the physics concepts led aggressive-confident students to
initiate group discussions and handle equipment more often than their peers. However,
they expected fellow group members to behave similarly and they expressed some dismay
at the apparent lack o f participation:
I t ’s not even a race, because you all leave at the same time. ...I think the people
that are maybe slower at it, they fe e l like, in the past, they’ve always had to keep
up, so they ’re trying to add as much as they can, so when they come up with the
answer, they ju st have to fill in the blank.
Aggressive-confident student
16
I suppose I learn better ...if I'm talking or I 'm active, when I 'm doing stuff with
my hands, or i f I ’m actually measuring. A nd in most o f my groups, everyone
says, “Well, here, you do it. I don't want to do it. ’’ They ’ll push, it to someone
other than themselves.
Aggressive-confident student
Unfortunately, attempts to motivate other students often did not prove successful. One
interviewee describes her efforts to increase group discussion:
I ’d usually try and make little comments, like, “What are you writing? ” or
“What do you guys think? ” or “We should talk about this, ” or “Do you guys
want to go over it? ” You know, ju st kind o f subtle little hints.
Aggressive-confident student
However, another member from the same group saw the prior student’s questioning as
hurried, answer-driven, and unproductive:
[S]he ’ll go so fa st that she ’ll get o ff on her own thing. A nd she ’ll be like, “Well,
did you get the same thing I did? ” A nd I'm like, “Well, I'm still a couple pages
behind you there. I ’ll tell you when I get there. ” And she does so much thinking
out loud that I ju st kind o f block her out...
Cooperative-uncertain student
It would seem that the motivations and actions o f aggressive-confident students were at
cross-purposes; in their attempts to generate greater participation among group members,
their behavior often frustrated their peers or was misinterpreted.
Cooperative-confident We classified students as “cooperative” if they seemed to
be comfortable in group settings. Unlike aggressive-confidents, “cooperative-confident”
students accommodated themselves to any pace; they could keep up with previouslyknowledgeable and aggressive-confident students, but they were willing to slow down for
less confident group members. These students used their comfort with the material to
move the group forward, but they dealt with less able students more magnanimously than
17
their more aggressive counterparts. While they still functioned as leaders in some cases,
individual participation increased in groups lead by cooperative-confident students:
Everyone discussed and put input in, and then there was RC and BC [both
cooperative confident students] that mainly discussed, and me and LA followed
somewhat, and then we set up things, and it was, everybody was putting
everything in kind o f equal; everybody had an equal share o f what was going on.
When there was time to write an answer, everyone would discuss what the answer
was, so everybody would be clear what it was, and then we 'd all write our answer
down.
Passive-uncertain student
The behavior o f cooperative-confidents depended largely on who they were grouped with.
They easily kept up with faster paced students. On the other hand, they were not afraid to
speak up when problems arose. For instance, one cooperative-confident described her
third group experience, which included a previously-knowledgeable student:
[This group was] a lot more fast-paced than my other two groups, but that’s okay,
because I fe e l comfortable enough to say, “I don’t know. We need to stop this
and someone explain it, ” or “L e t’s talk it through, and tell me how you think you
got it this way. ”
Cooperative-confident student
W hen grouped with more unsure students, cooperative-confidents saw themselves
as teachers or motivators. Unlike the aggressive-confidents, however, they did not seem
to classify themselves as “better” students and their attempts at helping were often more
successful:
D M works slower than all o f us, and A S [a cooperative-confident] and I, we help
her o u ta lot as fa r as helping her figure out which way she needs to go, not by
giving her the answers but by asking her questions to make her think o f where she
needs to be.
Cooperative-uncertain student
18
Cooperative-uncertain** These students were very similar to cooperativeconfidents in their work style and understanding o f the material. However, they differed in
their level o f participation. While cooperative-confidents were comfortable speaking up
and even controlling groupwork, “cooperative-uncertain” students rarely initiated
discussion and relied on more outgoing group members to seek them out. This meant that
cooperative-uncertains could be left behind in group discussions and sometimes had
trouble with more complex topics:
[VJoliage confused me, so I ju st kind o f learned on that one, because I didn't
really understand it. Occasionally, i f I did understand it, I could help the other
ones in the group that didn't, but that one I didn’t really understand so I ju st
learnedfrom the group.
Cooperative-uncertain student
She [a passive uncertain] let other people run the show, and I did to some extent
too. They wanted to take on the leadership role like that, and I wanted to be
involved and so forth, but I didn 't want to do it my way all the time.... A t times,
we 'dfe e l a little bit left behind too, some o f us; w e’d still be writing our data,
[and they ’d be] getting the experiment ready, going on ahead.
Cooperative-uncertain student
In spite o f their decreased participation, cooperative-uncertain students worked very
effectively one-on-one with passive-uncertain students. They were comfortable acting as
mentors and offering encouragement when not the focus o f the entire group. In several
instances, a cooperative-uncertain student served as a liaison between a passive-uncertain
student and the rest o f the group:
There was one girl [a cooperative-uncertain], I think she, like, could have stayed
with us, but she helped the other one [a passive-uncertain] a little more because
she knew she was a little slower and look a lot more time to figure it out.
Cooperative-confident student*
** “Uncertain” refers to the students’ self-perceived group roles. Uncertain students were less comfortable
initiating group discussions and tended to isolate themselves from more outgoing group members.
19
He [a cooperative-uncertain] was like the people that was [sic] in the first group.
[in which she had a positive experience], because h e 'd say, “D idyou
understand? ” and I 'd say, “Yeah, ” and i f I didn't, I 'd ask him. I fe lt really
comfortable working with him.
Passive-uncertain student
I think I 'm kind o f more o f a teacher to D M [a passive uncertain student]. I f
she ’ll be working on homework and she doesn ’t get it, I ’ll work with her after
class, to figure it out and help her with it.
Cooperative-uncertain student
While cooperative-uncertain students were not active group participants, they
were better than passive-uncertains about voicing complaints or opinions, especially in
relation to focus or group pace:
[S]he ’ll ramble on about what she thinks the answer is, and then she ’ll be like,
“Well, what did you get? ” “Well, I ’ll tell you when I get there. ’’ I ju st kind o f
block her out until I ’ve done the work to where I ’m happy with what I ’ve g o t
Cooperative-uncertain student
W e’d do an experiment, and often times each o f the individuals would write down
their own conclusions without sharing them, so I would try to slow the project
down that way and ask them some questions about it...
Cooperative-uncertain student
Cooperative-uncertain students were aware o f their limits and were less willing than most
to be rushed through an activity.
Passive-uncertain Students who exhibited quiet and withdrawn group behavior
were considered “passive.” These students might develop a relationship with a single
student within a group, but spoke very little and spent most o f their time writing or
observing. “Passive-uncertain” students were the least comfortable working in groups.
Although m ost ultimately developed an understanding o f the material, they avoided
20
contact with the equipment and had to be dragged almost forcibly into discussions. The
one passive-uncertain student interviewed described her quiet behavior as ingrained:
I should learn to speak out more, but I ’ve ju st been raised where speaking out
isn ’t right, that i t ’s wrong ...
Passive-uncertain student
Other students felt this behavior stemmed from a lack o f self-confidence, and expressed
frustration at such passiveness:
Some people do n ’t like to do stuff in fear o f messing it up...
Previously-knowledgeable student
We wanted to at least make sure she would participate... I actually ran into her
and said, “No, ju st go fo r it. Who cares i f i t ’s dumb? Just, you know, throw it
out there, and we ’ll talk it through. It doesn’t matter. ” You ju st have to
participate. No question is a dumb question.
Cooperative-confident student
Passive-uncertain students also seemed to require additional time to learn concepts. The
behavior o f passive-uncertain students led to faster, more vocal students creating upsetting
stereotypes. In one instance, a previously-knowledgeable student saw quietness as
negative and related it to a lack o f understanding:
There’s quiet people in the group, and i f y o u ’ve got a loud person, that’s going to
draw them out. That’s better fo r the quiet person. And the loud person is
practicing skills o f communication, going, “She’s really quiet. I need to make
sure sh e’s understanding this. ”
Previously-knowledgeable student .
But, when a passive-uncertain student describes a scenario similar to the one above:
ft It should be noted that this student was of Native American descent, and her passive nature appears to
stem from a strong cultural background. Future studies w ill offer a broader sample o f self-reports from
passive-uncertain students.
ND really took on a dominant role. She fe lt like she knew everything. She would
explain concepts to us; it was like she would talk down to us. A nd she wouldn't
give us time to discuss...
Passive-uncertain student
Such student reactions serve as an important reminder o f the distinction between ability
and personality.
Problems Uncovered bv the Interviews
Although the interactions among these five types o f students led to several
successful groupings, these were overwhelmed by students’ negative experiences. Two
broad categories emerged from analysis o f the interview transcripts:
Group Pace
Virtually every student complained about the pace o f at least one group. In fact,
the instructor noted that all groups appeared to w ork fairly evenly. Tasks were assigned
for each day, and only rarely did a group fail to finish within the tw o-hour time limit. This
is not to say that all groups finished simultaneously; typically, a group would take between
75 and 120 minutes to complete a module. However, even within this narrow time frame,
students felt pressured to w ork at a pace with which they were uncomfortable.
M ost students complained o f feeling rushed. Their complaints were consistently
directed towards previously-knowledgeable or aggressive-confident students:
[T]hey moved very fa st and I ju st kind o f tried to keep up with them... just,
“Come on, let's keep going. ” Wewere usually finished an hour before the rest o f
the class every time, and that was a hard pace fo r me to keep.
Aggressive-confident student discussing two
previously-knowledgeable students
It can be frustrating sometimes, because BA will get done on her own so quickly,
and sh e's so loud, and she ’ll keep “What 'dyou get? What'dyou get? What'd
you get? ” and I'm still behind her, and she doesn't waitfo r us.
Cooperative-uncertain student discussing an
aggressive-confident student
I wanted more time to explore experimentally a bit more in depth, and we didn't
get to do that at times, !w anted a bit more time to actually look at things, instead
o f coming in, performing it, writing down the numbers, and getting out o f class.
Cooperative-uncertain student
M ost o f the previously-knowledgeable and aggressive-confident students accepted a
slackened pace, albeit unhappily. Unfortunately, their groupmates seemed to sense this
hostility, and this often led to tension within the group:
Itfee ls like, i f I could go at my own pace, I ’d go, go, go and get done a lot faster.
Aggressive-confident student .
[A J lot o f the time I think he thought, “Gee, this is pretty simple and I wish I
couldjust get it done and leave, ” and the rest o f us were like, “Well, we d o n 't
really know it, so we want to go a little slower. ”
Cooperative-uncertain student discussing a
previously-knowledgeable student
In Some cases, a pace dichotomy formed, either o f ah individual functioning
separately from the rest o f the group, or the group splitting into tw o pairs:
There’s one gal in there [apreviously-knowledgeable student]... sh e'spretty
■quick with things too. She goes ahead quite a bit, and often times myself and
another gal in there, PM, feel that we ’re always playing a catch-up game.
Cooperative-uncertain student
(
23
STUDENT: [W]e work together, but we work separately at the same time. We ’ll
do the work, andthertwe Hl Jdndofget together, and see who got what, and i f
we ’re the same, then we ’re happy, we go on, and i f we ’re different, then we ’ll
break it down to see where [the problem is]. We ’ll all help each other out along
the way.
INTERVIEWER: I ’m getting the impression that it doesn 't always operate as a
group. You do things separately and then you just talk fo r a few minutes.
STUDENT: Uh-huh. Real separate.
Cooperative-uncertain student
Two pairs working separately produced more positive results than the psychological tugof-war produced in groups where a single member forged ahead. However, the
instructors had hoped to develop successful groups o f four, and these groups failed to
accomplish that.
Extrem e Heterogeneity
The w ork pace o f a given student related to how quickly he or she grasped the
material. Unfortunately, most students took w ork pace as a proxy for ability; faster
students were perceived to be more intelligent and vice versa. Previously-knowledgeable
students expressed frustration with slower group members, but they attributed a slackened
w ork speed to laziness or lack o f intelligence:
We [two previously knowledgeable students] would really fly through it, because
we really helped each other out a lot. The girl that liked to ask the teacher the
answers to questions, she liked to stop us a lot, just fo r the fa c t that she didn H
really want to think fo r herself...
Previously-knowledgeable student
In one glaring example o f the dichotomy that developed between slower- and
faster-paced group members, two students - one previously-knowledgeable and one
passive-uncertain - described an incident where they worked on an assignment together at
24
the end o f a class period. The instructor observed conflict between the Studentsi and after
the encounter, the passive-uncertain student approached the instructor, unconvinced o f the
methodology used by the previously-knowledgeable student. It was discovered that the
passive-uncertain student had an accurate understanding o f the material and the
previously-knowledgeable student had completed the assignment incorrectly.
In her interview, the previously-knowledgeable student recounted the incident.
She did not recall that her work was incorrect. Instead, she remembered her disagreement
with the passive-uncertain student as stemming from a deficiency on her group member’s
part:
One o f the girls in the. group, Iguessshe has not had a great math background,
and most o f the members o f my group left, and I stayed there to make sure she
understood what was going on... And you know, i f she doesn't understand the
math, she needs help o f some sort. A nd I was willing to do that, but there are
times when you ju st go, “Oh! H ere’s a calculator! Go! ” But i t ’s a matter of, I
d o n 't think she had the math skills to be able to do this, or the math learning
skills inside her head to be able to do some o f this stuff quick like the rest o f the
group.
Previously-knowledgeable student
It never occurred to the previously-knowledgeable student that she herself could be
incorrect.
Such excessive self-assurance unfortunately led to decreased cooperative learning.
Two scenarios emerged. In one case, a previously-knowledgeable student would see himor herself as a “teacher:”
They looked to me going, “Oh, hey, you have experience in this. ” And I ’m going,
“Okay. ” So that was interesting because I could teach, and I could learn some
more to make sure my past knowledge was what it should be.
Previously knowledgeable student
In groups which are organized to produce an end product - a report or a problem set such “group teaching” is usually a goal o f the instructor. In the case o f inquiry-based
learning, however, instructors wish to achieve maximum participation on the part o f all
group members and such “teaching” is generally viewed as detrimental.
Remembering that the present course is designed for pre-service elementary school
teachers, it could be argued that, for this particular group o f students, self-teaching would
be an acceptable approach. Unfortunately, the “teaching” techniques used by previouslyknowledgeable students were suspect:
[IJfshe didn ’t understand, !would, you know, “Here, you need to do this and you
need to do this. You get this number so you can do this with it. ”
Previously-knowledgeable student
This previously-knowledgeable student has failed to make a distinction between teaching a
concept and giving an answer. In other cases, previously-knowledgeable students had
trouble communicating with their fellow students, who expressed frustration:
[H]e knew how it would go, and we were like, “Maybe i t ’s supposed to go this
way, ” and he 'd try and explain. I think he, like, knew more than what you need to
know fo r the class. Like most o f the people in the class haven’t really had much
physics, and I think he had, so he was really more advanced, and so it was harder
fo r him to explain it, because he almost knew too much.
Cooperative-uncertain student
Ifw e were trying to understand something, instead o f just trying to explain it,
sh e 'd try and teach it.... And there was [sic] times when we started a new
concept, and it was really basic, so she 'd try and make it harder than it was...
She was really hard to deal with.
Passive-uncertain student
In spite o f the fact that the students were all future teachers, previously-knowledgeable
students unilaterally failed to demonstrate productive teaching methods.
Alternatively, previously-knowledgeable students who weren’t trying to function
as instructors tended to dissociate from the groups altogether:
We all worked, us three, and then HC [a previously-knowledgeable student] was
kind o f over there, doing his own thing.
Aggressive-confident student
Since students were graded on the success o f their group dynamic, with all students in a
group receiving the same score, this became a source o f genuine concern, albeit one that
presented no easy solution:
I think he really didn ’t like the group thing. He kind o f said that now and then,
’cause you know you get graded on your group, and when he would move ahead,
he would get a lower grade fo r his group, but i t ’s really not his fault, because he
ju st knew more, and so I think it was harder fo r him, ’cause i t 's hard, when you
know a lot more than the other people, to slow down.
Cooperative-uncertain student
In a single instance, a group managed to incorporate a previously-knowledgeable student
by what appeared to be sheer force o f will on the part o f a cooperative-confident
student9 . She focused on their future roles as teachers:
[BJecause when we teach elementary school kids, we ’re not ju st going to say,
“This is the answer. ” There’s got to be a process.
Cooperative-confident student
At the other end o f the spectrum, more advanced students (not necessarily '
previously-knowledgeable) occasionally expressed frustration at working with passive-
# -phis particular group contained a previously-knowledgeable student, a cooperative-confident student,
and a cooperative-uncertain student. It should be noted that, while the previously-knowledgeable student
was a traditional-age college student, both remaining group members were returning students above
traditional age. Our follow-up study (see Chapter 4) finds that in some cases, students above traditional .
age function similarly to previously-knowledgeable students, and provides a reasonable explanation for
the successful dynamic o f this group.
uncertain students. Complaints stemmed from an observed lack o f confidence and the
resulting diminished participation:
P d say, “Oh, \vell, why d on’tyou do it this time? ”... because, it would be a lot o f
“Well, what do you think? ” and it would drive my nuts. “Well, okay, what do you
think it is? ” you know, so that some people would participate more and give an
answer. That was pretty frustrating, and it lead to a lot of, “L e t’s take a break. ”
Cooperative-confident student
In some cases, the passive-uncertain students were genuinely more comfortable letting
more confident students handle the equipment:
I wasn 't very comfortable with the setting up when we did voltage. I had missed a
day..., but there was setting up the circuit using the amp and the volt. I w asn't
very comfortable about it, but him and ND would set things up.
Passive-uncertain student
M ost passive-uncertains, however, did not see themselves as less capable. Rather,- they
felt pushed aside by faster students:
[S]he always wanted to be the one that measured, Hkefound out how much it
weighed. Like i f anyone else would do it, she'd like, “No, no, no, no, no, ” and
sh e ’d take over and do it but not really show you. I think she had more o f a math
background, so she 'd come up with all o f these equations, and we ’re like, “Hold
on, we do n 't really have to do that. You ’vejust got to weigh it. ”
Passive-uncertain student
One passive-uncertain student, relating a particularly negative group experience, said:
[I]n the third group, it ju st shut me down. I didn 't want to try because the person
was so dominant.
Passive-uncertain student
Ifth is is the sort o f response generated by students as a reaction to cooperative learning,
w e must seriously re-examine our approach as instructors and our criteria for successful
group composition.
Group Success Rate
A summary o f the group compositions and their success rates is shown in Table I.
O fth e fourteen groups examined in this study, five demonstrated a “successful” group
dynamic. A successful group was one about which interviewees generated positive
reports, characterized by a high degree o f participation on the part o f all group members.
All groups that lacked a hierarchy performed well, but several groups with a distinct group
leader still allowed the participation o f all "members. In these situations, the leader
functioned more as an organizer than a source for information. Successful groups seemed
to draw from more limited, adjacent classifications, and most o f them lacked an
aggressive-confident or a previously-knowledgeable member. It is interesting to note that
all but one successful group§§ contained students from consecutive student classifications
(for example, a passive-uncertain is not grouped with a cooperative-confident without also
being grouped with a cooperative-uncertain). This probably stems from the tendency o f
slightly more confident students to act as “mediators” between the group and less
confident students (as demonstrated specifically with cooperative- and passive-uncertains).
Two groups were labeled “partially successful.” These were groups in which three
o f the members worked well together and produced a productive and cooperative
•
dynamic. The fourth member voluntarily withdrew from group participation. In both
these cases, the outside group members were previously-knowledgeable students, and no
negative effects could be found from their lack o f participation.
H This is the group mentioned in the previous footnote. See Chapter IV for more discussion.
29
Successfiil
CC, CU (x 2), PU
CC (x 2), CU, PU
CC, CU (x 2)
AC, CC (x 2), CU
PK, CC, CU
Partially Successful
AC, CC, CU, PK
CC, CU, PU, PK
Unsuccessful
AC, CU (x 2), PU
PK, CC, CU, PU
PK (x 2), AC
PK, AC, CU, PU
CC, PU (x 3)
AC, CC, CU, PU
PK CU (x 3)
TA B LE I: A breakdown o f group composition in relation to group success. The student
categorizations have been abbreviated as follows:
PK - previously-knowledgeable
AC - aggressive-confident
CC - cooperative-confident
CU - cooperative-uncertain
PU - passive-uncertain
M ost o f the groups are four-member groups, but three groups consisted o f only three
students. Students in italics indicate scenarios where three o f the four members functioned
as a successful group, and the fourth member worked individually, with no real impact on
the group.
A group was considered unsuccessful if students reported predominantly negative
experiences about it, or if group members were unable to function in a reasonably
cooperative manner. The seven remaining groups fit these criteria. Problems within these
groups were ultimately derived from extreme group heterogeneity. The most pronounced
difficulties involved previously-knowledgeable students - either a previouslyknowledgeable student tried to assume “command” or worked at a pace that was
intolerable for most o f the other group members. Even without previously-knowledgeable
students, problems developed between faster- and slower-paced students; no successful
groupings were observed that placed aggressive-confident students with passive-uncertain
students. Here, problems manifested themselves mostly as pace dichotomies within a
group. In one case, outright conflict resulted.
In two cases, groups were created in which three students were in the same
“uncertain” classification and the fourth student was in a more “confident” classification.
These groups were formed with the intent that the more confident student would serve as
a natural leader. In one case, the group faltered because the three passive-uncertain
students failed to participate, leaving the cooperative-confident student to carry the group
single-handedly. In the second case, the previously-knowledgeable student refused to let
the three cooperative-uncertain students participate adequately and assumed an almost
dictatorial role.
I
W e were puzzled that the groups that demonstrated the most difficulty seemed to
be o f the type most highly advocated by conventional literature recommendations extremely diverse groups, in which.higher ability students have the. opportunity to “teach”
those o f lower ability. In our next section, we examine the justification for previous
conclusions on group learning, and see how well those goals match our own.
CHAPTER 3
OUR RESULTS VERSUS RESULTS OF PREVIOUS STUDIES
Previous research on collaborative small-group learning has typically produced
results different from those found in this study. Many previous studies (Cottle & Lunford,
1995; Emkey31979; Heller & Hollabaugh31992; Heller, Keith3& Anderson31992; Lunetta3
1990; Miller, Wilkes, & Cheetham3 1993; Peterson et a l, 1984; Rau & Heyl3.1990; and
Webb3 1985, for example) have examined similar examples o f group learning. A typical
study looked at an introductory physics courses where students, working in groups o f
three to four, were presented assignments that they were meant to approach as a group.
As with our Physics by Inquiry course, instructor participation was mostly through
Socratic dialogue. Several variations o f student grouping were used. Cottle & Lunford
(1995) allowed students to form their own groups for each class period in which
cooperative groups were used. There was no instructor input, and groups were not
expected to stay together from one class period to the next (although no evidence is
presented as to whether students actually rotated groups or if they mostly remained in the
same group throughout the course). Two studies (Heller, Keith, & Anderson, 1992;
Miller, Wilkes, & Cheetham, 1993) rotated groups several times each quarter, allowing
the first round o f groups to form randomly and then forming subsequent groups. Heller,
Keith, & Anderson (1992) placed students into academically heterogeneous groups, and
32
Miller, Wilkes, & Cheetham (1993) first organized homogenous ability groups before
proceeding to heterogeneity. The remaining studies carefully organized students into their .
groups from the beginning. In most cases, these groups were composed o f students
displaying maximal academic heterogeneity (Emkey, 1979; Heller & Hollabaugh, 1992;
Rau & Heyl, 1990); however, Webb (1985) compared groups o f heterogeneous ability to
more uniform groups.
M ost o f these studies concluded that a cooperative group environment was
beneficial to all o f the students involved. Furthermore, studies that examined group
composition argued that groupings o f maximal heterogeneity demonstrated the strongest
“positive” results. The consensus interpretation for collaborative group learning
advocated by these investigators is that more advanced students will assist novice group
members with difficult concepts. The truism is that the more advanced student is forced
to organize and clarify his or her knowledge by teaching, while the novice student benefits
from an additional source o f information. As Lunetta (1990) states: “The less-skilled
student profits from the insights and explanations o f the more talented student. The more
talented student can enhance learning by talking through the material with others and
responding with questions.” 1 All students come away enriched.
Our results, in contrast, strongly suggest that maximal heterogeneity was actually
related to the greatest .problems in this type o f learning environment. Moreover, it appears
that the most advanced students were even detrimental to the overall group
1 V. N. Lunetta, “Cooperative Learning in Science, Mathematics, and Computer Problem Solving,” in
T ow ards a S cien tific P ra ctice o f Science E ducation, ed. M. Gardner (Hillsdale: Lawrence Erlbaum •
Associates, Inc., 1990), 239.
learning experience, hindering both the learning o f more novice students and failing to
advance their own understanding. In this chapter, we highlight how our data differs from
traditional norms and suggest some causes based on interviews.
. Reviewing Comparable Physics Education Group-work Research
Attempts to study group learning tend to start from the assumption that interaction
with other classmates will draw students into the learning process and ultimately allow
students to internalize more o f the course material than they might in a more detached
setting, such as a lecture environment (Emkey, 1979). There is also the hope that active
discussions and encouraging cooperation will improve problem solving skills (Heller,
Keith, & Anderson, 1992; Heller & Hollabaugh, 1992).
Many group learning efforts start with three basic premises: I) maximal
heterogeneity provides the most productive work environment, 2) the learning o f group
members can be determined by comparing the highest score o f an individual in the group
to the score given to the group as a whole, and 3) group success can be adequately
assessed on purely academic grounds This assessment is easy to administer widely and
definite results can be obtained without the same time expenditures that might be required
in using focus groups or interviews. However, reviewed research studies tend to ignore
the assumptions behind these premises.
Consider the justification for maximal heterogeneity. M ost studies cite Johnson et
al. (1984), who state: “M ore elaborative thinking, more frequent giving and receiving o f
explanations, and greater perspective in discussing material seems to occur in
34
heterogeneous groups...”2 N o evidence from specific research results is offered to
support this statement. Instead, their academic justification for heterogeneity is based on
the claim:
“It is obvious that low- and middle-achieving students have much to gain from
working in cooperative learning groups with high-achieving peers. In terms o f
motivation and actual achievement, the largest gainers from working in
heterogeneous cooperative learning groups are the struggling, low-achieving; the
next largest gainers are the middle-achievers.” [underline added]3
It is interesting to note that the idea o f heterogeneous grouping has become so
ingrained that its validity as an effective teaching strategy is rarely questioned. As Good
& Marshall (1984) state:
The heterogeneous-homogeneous question [is] typically approached as a onevariable problem. This is, much o f the research is based on the belief that there is a
single answer to the question, Is homogeneous or heterogeneous grouping better?
This belief has probably led many investigators to study class heterogeneity
without carefully considering the types o f students involved, the community being
served, the quality o f instruction, and many other variables.4
In fact, it is this consideration o f other variables which has generally been neglected in
physics education research. M ost o f the studies we considered either started with the
premise that heterogeneous groupings were preferable, or they used assessment strategies
that did not directly reflect group dynamics (discussed on pages 38 and 39). However, in
studies which make more direct observations o f internal group functions, investigators
have found that heterogeneous groupings are not necessarily the most favorable. Good
2D. W. Johnson and others, Circles o f Learning: Cooperation in the Classroom (Edina: Interaction,
1984), 27.
3 Ibid, 74.
4 T. L. Good and S. Marshall, “D o Students Leam More in Heterogeneous or Homogeneous Groups?” in
The Social Context o f Instruction: Group Organization and Group Processes, eds. P. L. Peterson, L. C.
Wilkinson, and M. Hallinan (Orlando: Academic Press, Inc., 1984), 17-18.
35
and Marshall (1984) give evidence that more “homogeneous grouping increased
participation and involvement in academic and social activities for most students,
especially slow learners.”5 Similarly, Webb and Kenderski (1984) observed that
“(s]tudents in high-medium or medium-low groups [showed the most interaction], on the
average, and students in high-medium-low groups [showed the least interaction], on the
average.”6 These results suggest that extremely heterogeneous groups may be less
productive and positive group learning environments than more homogeneous ones.
The second premise, that the learning o f individuals within the group can be
determined by assessing group achievement, makes some very broad tacit assumptions.
Johnson et al. (1984), like many subsequent studies (Cottle & Lunford, 1995; Heller &
Hollabaugh, 1992; Heller, Keith, & Anderson, 1992, etc.), seem to imply that the work
produced by the group accurately reflects the retained individual knowledge o f each
student in the group, and that all group members are able to reach the same level o f
achievement as the highest ability students within the group. For example, Heller &
Hollabaugh (1992) state that, “[a]n examination o f written problem solutions indicated •.
that instructor-assigned groups o f mixed ability (e.g., a high, medium, and low ability
student) performed as well as groups consisting o f only high-ability students... ”7
5 Ibid, 20.
6 N. M. Webb and C. M. Kenderski, “Student Interaction and Learning in Small-Group and Whole-Class
Settings,” in The Social Context o f Instruction: Group Organization and Group Processes, eds. P. L.
Peterson, L. C. Wilkinson, and M. Hallinan (Orlando: Academic Press, Inc., 1984), 162.
7 P. Heller and M. Hollabaugh, “Teaching Problem Solving through Cooperative Grouping. Part 2:
D esigning Problems and Structuring Groups,” American Journal o f Physics 60 (July 1992): 641.
36
However, these same studies base their results on outcomes produced by the group as a
whole. The group assessment techniques used say nothing about individual student
understanding. The assumption seems to exist that, as a result o f their group experiences,
all come away with comparable grasps o f the material.
I f w e reexamine the hypotheses researchers pose about interactions within such
heterogeneous groups, this reasoning may be inconsistent. As Johnson, Johnson, & Smith
(1992) state that:
. ..o ra l discussion o f relevant information has at least two dimensions: oral
explanation and listening. Both benefit the giver and receiver. The giver benefits
from the cognitive organizing and processing, higher level reasoning, insights, and
personal commitment to achieving the group’s goals derived from orally
explaining, elaborating, and summarizing information and teaching one’s .
knowledge to others. The receiver benefits from the opportunity to utilize
others’ resources in their goal accomplishment efforts.8
The Johnson, Johnson, & Smith (1992) hypotheses o f interactions between
“givers” and “receivers” are borne out by Heller & Hollabaugh (1992), who observe that
“the higher-ability student typically supplied the leadership by generating new ideas and
approaches to the problem, [and] the low or medium ability student kept the group on
track by pointing out obvious, simple ideas.. .”9 While explaining may help solidify
understanding, it seems unreasonable to expect that the “giver” can raise the “receiver” to
a comparable level o f cognition, since the “receiver’s” learning is largely second-hand.
W hen w e consider a further statement by Johnson, Johnson & Smith (1992), it becomes
8 D. W. Johnson, R. T. John, and K. A. Smith, A c tiv e Learning: C ooperation in the C ollege C lassroom ,
(Edina: Interaction, 1992), 6.
9 Heller and Hollabaugh, 641.
37
difficult to justify in what context a consistent “giver-receiver” relationship can constitute
effective group learning:
The critical issue in understanding the relationship between cooperation and
achievement is specifying the variables that mediate the relationship. Simply
placing students in groups and telling them to work together does not in and o f
itself promote higher achievement. It is only under certain conditions that group
efforts may be expected to be more productive than individual efforts. Those
conditions are clearly perceived positive interdependence, considerable promotive
(face-to-face) interaction, felt personal responsibility (individual accountability) to
achieve the group’s goals, frequent use o f relevant interpersonal and small groups
skills, and periodic and regular group processing.10
I f students are placed in groups where higher ability students are expected to
function as teachers or helpers to lower ability students, it seems that many o f these
expected behaviors (for instance, positive interdependence or personal accountability) may
fail to materialize, in light o f the unequal status that group members may perceive amongst
themselves. The supposed “contributions” by low? and middle-ability group members may
not be the independent thinking that Heller & Hollabaugh (1992) assume. Webb (1982)
has evidence that “less proficient members [tend] to conform to the judgment o f more
proficient members, regardless o f the quality o f their judgments.” 11 This suggests that
higher ability students may possess an extremely dominant intellectual role, and that the
lower ability students are more likely to be receivers than contributors o f knowledge.
There is even evidence that group members are more likely to defer to perceived
rather than actual proficiency. Steiner (1972) describes a study in which B-26 bomber
10 Johnson, Johnson, and Smith, 14.
11 N. M. Webb, “Student Interaction and Learning in Small Groups,” R eview o f E d u ca tio n a lR esea rch 52
(Fall 1982): 432.
38
crews were presented with problems in cooperative group situations.12 Pilots, which
tended to have the highest status within crews, were the most successful in swaying
groups to their opinions, regardless o f how correct their answers were. On the other
hand, gunners, which tended to have the lowest status, were least successful. They also
had the most difficulty swaying groups to their opinions when they were correct. The
experience Steiner (1972) describes is not unlike the conflict discussed on pages 23 and
24, in which a previously-knowledgeable student with an incorrect understanding o f a
problem refused to even consider the correct explanation o f a passive-uncertain. Both o f
these examples suggest that the dominance higher ability students experience in
heterogeneous groups may not only encourage lower ability students to function as
passive learners, but. if higher ability students possess an incorrect understanding o f the
subject matter, they may be more likely to influence their groupmates. Conversely, lower
ability students with correct views may not only have problems convincing their fellow
group members, but may actually be dissuaded from these views by higher ability students.
Neither situation is desirable.
N one o f the physics education studies clearly demonstrate that w ork in groups o f
maximal heterogeneity ultimately leads to gains for all students. This would require
assessment o f individual students grouped in heterogeneous groups in relation to a control
o f students who learned by a more traditional method. Peterson et al. (1984) did perform
such a study, and found a significant negative correlation between performance on
achievement tests and receiving explanations in groups. “One obvious hypothesis,”
121. D. Steiner, G roup P ro cesses a n d P roductivity, (New York: Academic Press, 1972), 25.
39
Peterson states, “is that low-ability students were being given the answers by others in the
group, and they wrote down the answers without having a clear understanding o f how to
w ork the problems themselves.”
Such a result might be overlooked by studies that
considered a final group outcome as the only evidence o f student gains.
The third premise - that assessment o f a group project provides evidence o f group
success - is even harder to justify. Beneath this assumption lies the idea that group
success can be measured academically - a group which solves problems correctly or
produces a report which receives a high grade has therefore functioned successfully - and
that high-scoring group outcomes can only be produced by groups that interact
successfully. M ost studies ultimately take this “end justifies the means” approach. Heller,
Keith, & Anderson (1992) assess their groups solely on outcomes, with the justification
that:
“In well-functioning cooperative groups, students can share conceptual and
procedural knowledge and argument roles, and request clarification, justification,
and elaboration from one another, so a better solution emerges than could be
achieved by individuals working alone. The results o f this study suggest that this
type o f collaboration did occur.” 14
While a successful group will function as described'above, no evidence is offered that a
group with a poor dynamic will produce outcomes which score poorly. In fact, in thenstudy o f group dynamics, Miller, Wilkes, & Cheetham (1993) found that “ [tjhere is no
correlation between group performance and group conflict, or between group *146
13 P. L. Peterson and others, “Merging the Process-Product and the Sociolinguistic Paradigms: Research
on Small-Group Processes,” in The S ocial C ontext o f Instruction: Group O rganization a n d G roup
P ro cesses, eds. P. L. Peterson, L. C. Wilkinson, and M. Hallinan (Orlando: Academic Press, Inc., 1984),
146.
performance and group satisfaction.” 1415 Although the focus o f our study does not consider
academic criteria in general, our data also found no appreciable difference in academic
achievement between students who participated in successful groups and students who
participated in unsuccessful ones. This implies that, while outcomes-based assessment can
be a good judge o f group academic prowess, it is not necessarily reflective o f successful
group interactions.
It appears that the assumptions made by Heller, Keith & Anderson (1992) are
reflected in many group learning studies. They believe that a good group will not only
share knowledge but will also participate equally and enthusiastically. However, when
many studies describe actual group interactions, myriad scenarios emerge which did not
typify good group dynamics. In some cases, groups exhibited distinct hierarchies:
“ Some groups had a dominant student who railroaded the group into an approach
or problem solution. At the other end o f the personality spectrum, a timid student
would be reticent o f participating and often became the silent record-keeper for the
. g ro u p ...” 16
In other situations, students dissociated from the group entirely:
“[Sjome students do not interact with their peers. These individuals tend to work
alone and do not contribute their ideas to other members o f the group. Those who
w ork outside the group could not be identified based on their performance in the
course...” 17
14 P. Heller, K Keith, and S. Anderson, “Teaching Problem Solving through Cooperative Grouping. Part
I: Group versus Individual Problem Solving,” American Journal o f Physics 60 (July 1992): 635
15 J. E. Miller, J. Wilkes, andR . D. Cheetham, “Tradeoffs in Student Satisfaction: Is the ‘Perfect’ Course
and Illusion?” Journal on Excellence in College Teaching 4 (1993):43.
16 Heller and Hollabaugh, 641-642.
17W .L . Emkey, “A Small Group Approach to Introductory Physics,” American Journal o f Physics 47
(August 1979): 696.
41
“In groups with four members, one student was invariably left out o f the problem­
solving process. Sometimes this was the more timid student who was reticent to
ask for clarification. At other times, the person left out was the most
knowledgeable student who appeared to tire o f continually struggling to convince
the three other group members to try an approach, and resorted to solving the
problem alone.” 18
Finally, some groups suffered from a distinct lack o f cohesion, especially in situations
where the group’s main focus was problem-solving:
“[S]ome o f the groups do not seem to ‘gel.’ Here individuals tend to work by
themselves showing little or no participation within the group... .”19
To try to eliminate such problems. Rail & Heyl (1990), as well as Heller, Keith, &
Anderson (1992) and Johnson et al. (1984), advocate the assigning o f roles within a group
“to prevent more vocal, aggressive, or precocious students from dominating group
discussion.”20 In these studies, role rotation seemed to enhance group dynamics
somewhat. However, evidence presented by S kala, Slater, & Adams (1999) and Steiner
(1972) showed that role assignments may not necessarily serve the purpose intended by
Rau & Heyl (1990). Skala, Slater, & Adams (1999) conducted focus groups o f students
participating in group learning activities within a large-enrollment (200 students per
section) lecture-based course and found that, unless closely monitored, students often
deviated from their prescribed roles. Specifically, “students did not always rotate the
position o f leader. The student with the most dominant personality usually emerged as
18 Heller and Hollabaugh, 640.
19 Emkey, 696.
20 W. Ran and B. S. H e y l, “Humanizing the College Classroom: Collaborative Learning and Social
Organization among Students,” Teaching S ocio lo g y 18 (April 1990): 146.
42
the leader and remained the leader through the whole semester.”212Similarly, Steiner
(1972) describes a study in which students were given personality tests and students were
grouped in pairs in which one student had a more dominant personality than the other.
Pairs were then given a task, and roles were assigned such that one student was given a
more dominant, directorial position. When the role o f director was given to the less
dominant student, pairs experienced a failure rate 27% higher than when more dominant
students were assigned to direct, and 18% higher than control groups in which no roles
were assigned.
22
Both o f these studies suggest that attempts to assign roles may be counter­
productive. N ot only is there evidence that students have difficulty conforming to roles especially students with dominant personalities - but findings show that role assignments
may lead to less productive interactions. In addition, students in our interviews described
that, in their own successful groups, group roles seemed to move naturally from person to
person, depending on the activity at hand:
STUDENT: [In my favorite group], some people would set up the equipment
right away...and anotherperson was interested in the questions in the text more,
and I always fe lt like I wanted to coordinate it a bit, get it all going.
But... sometimes other people would set up the equipment too and the person
usually doing it was accepting o f that. It was no big deal i f somebody else did it
too, so it was good that way...
INTERVIEWER: [It] was...fluid?
STUDENT: Yeah, I thought so.
21 C. Skala, T. F. Slater, and J. P. Adams, “Qualitative Analysis o f Collaborative Learning Groups in
Large-Enrollment Introductory Astronomy,” P ro ceed in g s o f the A stron om ical S o c ie ty o f A u stra lia (in
press): 6.
22 Steiner, .56. .
43
I f roles had been assigned, students would have been expected to function in a particular
way within the group for some arbitrary period o f time, and such intuitive role-taking and
switching might have been suppressed.
Towards the Implementation o f Successful Cooperative Groups in Inquiry-based !.earning
Upon carefully examining previous studies, we feel we possess a better
understanding o f observed differences between our conclusions and others. It is important
to remember that, while previous studies have focused on collaborative small-group,
learning, they have been assessing traditional - rather than inquiry-based - physics
courses. The physics courses in question have also typically been directed towards science
and engineering.maj ors, while Physics by Inquiry concerns itself with a unique audience the pre-service elementary school teacher. This creates two challenges that may not be
concerns in other courses: I) students will probably enter Physics by Inquiry with a far
greater range o f backgrounds, fears, and expectations, and 2) if the students come away
from the course with a less than positive and successful experience, they might.choose not
to incorporate similar activities into their own classrooms. The instructors are put in the
delicate position o f trying to challenge the students intellectually while ensuring their
success.
Therefore, rather than assess knowledge solely through exams or graded problem
sets, instructors work very closely with all students and assessment takes place
continuously. Groups are not allowed to advance more than a page or two in any one
module w ithout participating in a discussion with the instructor. With a student-to-
44
instructor ratio o f about 7-to-l, this environment also draws instructors much closer to the
group experience. Instructors are able to focus more closely on how students interact
with one another, but are also more acutely aware o f tension and lack o f productivity.
W here group learning is often used to supplement more didactic physics
instruction, the outcomes-based assessment used by cited studies might be entirely
adequate. Students who intend to focus on science and engineering at the college level
might be more likely to enter courses with similar backgrounds, eliminating the problems
presented by previously-knowledgeable students. If students are grouped to work
quantitative problems with a single correct solution, the “teacher-learner” relationship can
undoubtedly be more productive than students working individually. Unfortunately, when
all o f the learning takes place in small groups and there are many intellectual paths to a
correct understanding, as in our environment, we have found that the traditional
approaches to group learning can fail.
Therefore, for the remainder o f this study, we grouped tw o sections o f Physics by
Inquiry, from the Fall 1998 semester, drawing on the hypotheses developed in Chapter 2.
In order to implement our grouping strategy, we presented the students with a instrument
(discussed in the Appendix) on the first day o f class which allowed us to categorize them
Using the results o f this instrument, we then segregated previously-knowledgeable
students into separate groups which were not rotated into groups containing nonpreviously-knowledgeables. Because o f the advanced understanding previouslyknowledgeable students already possessed, we expected to move these students through a
different, slightly accelerated curriculum to keep them challenged.
After the previously-knowledgeable students were segregated, the remaining
students were organized in moderately heterogeneous groups, making sure to only group
adjacent categories and include no more than two students from one category. We
avoided grouping students in any way that we felt might produce negative outcomes. It
was felt that placing students in detrimental situations solely for the sake o f a control
group would be unacceptable teaching practice; therefore, groupings from the Spring
1998 semester were the closest we considered to a control group.
At the end o f the Fall 1998 semester, we conducted interviews similar to those
conducted with the Spring 1998 students. In this case, however, w e attempted to
interview all o f the students in both sections, providing us with thirty to forty interviews.
W e determined that our grouping strategy would be successful if w e saw a drop-off in
conflicts and pace dichotomy, and a distinct increase in student enjoyment and
participation. Our results are discussed in Chapter 4.
46
CHAPTER 4
IMPLEMENTATION AND FOLLOW-UP
To test the validity o f our conclusions, we used our results as a basis for formally
assigned group composition in both sections o f the.Fall 1998 Physics by Inquiry course.
A classification instrument (discussed in the Appendix) was created, which was
administered to students on the first day o f class. The instrument was designed to be
simple enough that previously-knowledgeable students could be identified immediately.*
Two out o f sixteen students in one section and eight out o f eighteen in the other were
identified as previously-knowledgeable.
Previously-knowledgeable students were grouped together exclusively, and an
accelerated curriculum was designed for these groups to accommodate the faster pace that
was expected to emerge. At the outset, all previously-knowledgeable students were
consulted individually about the implementation o f a dual-pace curriculum, and all agreed
to participate. Due to slight changes in the curriculum, students participated in three
groups during the course o f this semester rather than the four different groupSt used in the
Spring 1998 semester.
* It was important that previously-knowledgeable students be identified quickly, as they were asked to stay
after the first class meeting to obtain their permission to carry out the grouping experiment,
t In the first section, the two previously-knowledgeable students functioned as a single group throughout
the semester. These are the only students who did not participate in group reorganization.
47
The distribution o f student types identified by the categorization instrument in each
section is shown in Table 2.
Student Type
Number o f Students
Number o f Students
in Section One
in Section Two
2
Previously-knowledgeable
8
Aggressive-confident
2
3
Cooperative-confident
9
4
Cooperative-uncertain
3
3
O1*
Passive-uncertain
0
TA B LE 2: Distribution o f students in each section according to their cooperative group
learning categorization.
Groups were then arranged according to the following criteria:
1. Previously-knowledgeable students were grouped together and in separate
groups from other students.
,
2. Non-previously-knowledgeable students were grouped with people from
similar or adjacent categorizations.5
This strategy led to the twenty-eight combinations o f student learning groups shown in
Table 3. These groups were the focus for this portion o f the study. In all, 34 students
were impacted, and 3 1 o f these students were interviewed** concerning their group
experiences using a script similar to that used in the interviews o f the Spring 1998
i According to the responses given on the instrument, no student identified him- or herself as a passiveuncertain. While it may be that no passive-uncertains enrolled in the Fall 1998 sections o f Physics by
Inquiry, w e must acknowledge the possibility that the instrument was flawed. This is discussed in detail
in the Appendix.
5 By the time some later groups were formed, some cooperative-uncertain students had matured into
cooperative-confidents. This led to one grouping o f technically non-adjacent types.
** Two students (a cooperative confident from each section) withdrew from the course before the end o f
the semester, and one aggressive confident from Section Two was not interviewed. The behavior o f each
o f these students is therefore inferred from information provided by fellow group members.
48
students. As before, interviewers were not known to the students and had no contact with
them before or after the interviews.
SECTION ONE
Round One
PK(x2)n
Round Two
Round Three
CC(x2),.
CU(x2)
AC,
CC(x2), CU
AC,
CC(x2), CU
CC (x3)
AC, CC, CU
AC, CC(x3)
AC,
CC(x2), CU
AC, CC(x2)
CC(x2), CU
CC(x3)
CC(x2), CU
CC(x2), CU
SECTION TWO
CC(x2), CU
AC(x3)
PK(x4)
PK(x4)
Round Two
CC(x2),
CU(x2)
AC, CC, CU
AC, CC, CU
AC, CC, CU
PK(x4)
PK(x4)
Round Three
AC, CU(x2)
AC, CC, CU
AC, CC(x2)
PK(x4)
PK(x4)
Round One
TA B LE 3: Student organization in each o f their cooperative learning groups according to
their groups learning classification. Group learning types have been abbreviated according
to the same system used in Table I in Chapter 2.
The follow-up study differed from the original in that a larger number o f students
and groups were considered, and a larger percentage o f the students (91% versus 47%)
were interviewed, allowing for greater cross-correlation. In addition, the 16 interviewees
and 13 groups o f Section One were taught by an instructor other than the researcher, in a
deliberate attempt to remove instructor impact as a predictor.
ft The previously-knowledgeable students in Section One constituted a single group which lasted the
entire semester. For purposes o f the clarity in Table 3, they are only listed once.
Results o f the Follow-up Study
The follow-up study was designed to deal with the problems revealed in the initial
interviews that appeared to stem from extreme group heterogeneity, for example, lack o f
student participation or stress caused by diversity in group pace. By limiting heterogeneity
(and more specifically, segregating previously-knowledgeable students), the researcher
hoped to .create a group learning environment in which all group members contributed to
discussions and students were motivated to ask questions and. Ieam actively through their
group members, as opposed to using fellow group members as repositories o f knowledge,
or conversely, failing to interact productively with other students because they felt their
classmates lacked relevant knowledge. In light o f the fact that all class members were pre­
service elementary school teachers, the researcher felt that criteria for success should stem
from active engagement in groups and perceived self-fulfillment on the part o f the
students. Therefore, course grades were not taken into account** and, for our purposes, a
group was considered successful if most o f its members described positive and productive
interactions.
'•
Group Success Rates
Group success rates were determined by taking combinations o f each member’s
rating o f that group and determining an average. Student responses were divided into
H There is some debate over the reliability o f grading as representative o f successful group dynamics
anyway. For our Physics by Inquiry class, only a small percentage of the grade considered group
participation at all, and there was no component that was dependent on group problem solving. Most o f
the grading was determined by written individual criteria, such as homework assignments, essays, and
exams, and any effect o f group dynamics on grades is therefore difficult to determine.
three categories: positive (+), negative (-), or neutral (0).§§ Responses were interpreted as
positive if the student related his or her group experience as being good and described
behavior consistent with cooperation and active learning. Examples o f positive responses
include:
We all seem to, i f somebody’s having a problem with something, then another
person can help out, and i f they can't help that person, then another person can
think o f another way and help them out that way. So it was like different kinds o f
thinking; even though someone might understand all the ways, you know, they
could explain it to someone else so they could understand it. That was really
nice.
I fe lt valued. I fe lt like whatever I needed to ask or question or talk about was
worthwhile. I never fe lt like they thought I was behind or anything like that. It
was realty nice, ju st because we kind o f established a nice casual sort o f
atmosphere; whatever question you had, that was fine. We went over stuff to
make sure we all understood it. It was really nice in all o f them. I didn "tfeel too
different about any o f them, ju st working with different people. It was good.
Ifo n e o f us didn't get it, everybody else would stop and say, “Okay, this is how I
thought through it, and this how I did it, ” and pretty much everybody always had
an idea...
A response was considered negative if the student related extremely negative experiences
when interviewed. These tended to involve animosity towards other students, expressions
o f frustration, and the like. For instance,
Well, I know that I T liked to be in control o f the group and that kind o f frustrated
me personally, and I think TC also, and we just, you know, when we really didn ’t
know the answer, it was difficult fo r us to actually, like, come together andfigure
it out...
They would be kind o f lacking and lagging behind. W e'd spend too much time on
one thing or something and I ’d be... I know that at times I would get frustrated
andjust kind o f work ahead a little bit, and then come back when they were ready
to go ahead, and that was a little bit frustrating.
55 As w e had access to responses from a much larger percentage o f students, this method was slightly more
rigorous than that used to determine group success in the Spring 1998 semester.
51
They both completely understood what we were doing, and I was going a little
slower, but it wasn’t slowing them down at all. Theyju st kept going ahead o f me,
and I was ju st finding myselffalling behind and I was so uncomfortable with both
o f them. Like, our personalities didn 't seem to click. ...Ifelt ju st like an idiot,
’cause they were ju st like flying, and I ’d ask the two questions all the time and
they'd kind o f give me this answer, like, that was so obvious, ju st the answer, like,
not how they got to it, stu ff like that. So I found myself just skipping over sections
to get caught up with them. I couldn't wait to get out o f that.
STUDENT: It [the group] gives me a headache a lot o f the time.
INTERVIEWER: A nd why is that?
STUDENT: They talk about what they ’re going to do after class and "da-da-dada-da ” and then they ’re like, “Come on, we. ’ve got to get this done! ” A nd I ’m
like, “Hey, I ’m already passed you guys. ’’ You know, I work a lot on my own on
this one, ju st ‘cause they sit there and they talk and they talk and talk, and i f I try
and say, “Hey, you guys... ” they ’re like, “Hey, we ’re ju st talking fo r two
seconds. ’’ A nd then when you try. and do something, they ’re like, “No, no, no.
I t ’s not like that, ” and they never are willing to look at it from somebody else's
perspective.
A neutral response was characterized by more ambiguous or unclassifiable opinions. In
these cases, students described slightly negative situations that they tolerated but were
willing to accommodate, or spoke positively o f interactions that they knew did not
contribute to a positive learning environment. M ost students had an intuitive grasp o f
what constituted positive and productive group dynamics, and neutral responses seemed
to be characterized by attempts to rectify partially dysfunctional groups, or at least self-.
awareness o f what types o f behaviors would correct the problem:
I guess F N always kind o f stepped up and’d say, “This is the way i t ’s gonna work
out, ” and SS always questioned - she had to see everything fo r herself before
sh e’d believe that it was true - and, I don't know, ML and I ju st kind o f sat back.
I always had my ideas. I always fe lt like I knew what was going on. I didn’t
always voice my opinion, ju st because I ’m not real sure o f m yself all the time.
I t seemed everyday - well, not everyday, a couple o f days - that somebody would
move ahead, or two people would move ahead and two people would be behind.
[The problem was] ju st lack o f everybody staying together.
STUDENT: I fe lt that FH worked on her own a lot... I ’d look up and she'd be
ahead three sections, and then I ’d be behind and I would have to keep asking
questions about where we were...
INTERVIEWER: So did you ask her to slow down, or how did you handle that?
STUDENT: Well, !have to talk with her about it, ask her questions about it, and
that would make her slow down, and after that, she would realize she was fa r
ahead and stop.
STUDENT: [W]e, like, didn’t really work as a group. We worked, like, more
individually until we had a question...
INTERVIEWER: What do you mean? I ’m not sure I understand.
STUDENT: Like, we didn't really, like, work through the stu ff together; we kind
o f worked each o f us at our own pace, and i f any o f us had a question, w e’d ask
someone, like, “W hat'd you get fo r this one?”
I d o n ’t think it was quite the same [as the first group]. I think D U and I were
always on the same level, but I think KS was at a very different level, and we
never knew i f she really was ju st having a bad day, or i f that’s ju st how she was....
I didn ’ife e l like we [PMandD U j were rude to her [KS] in any way. I think she
fe lt closer to her first group, and I fe lt like she didn’t want to get to know us very
well.
6
I think she feels that... I don’t why she does, but sh e’s thinking she 'd like to do
most o f the work. I f [the instructor] comes over and there needs to be explaining,
you know, she ’ll speak up right away... A nd that’s kind o f hard, because... you
fe e l the tension there and you don’t really know what to do... I mean, just
because we ’re not jumping to speak up, it doesn’t mean that we ’re quiet ones and
we ’re not understanding or we need self-confidence or something.
Classifying neutral responses proved difficult. The ultimate criteria used was one o f
“acceptability:” whether the nature o f the group interactions was considered tolerable by
the researcher. It seems unlikely to expect that all group learning situations, no matter
how ideal the grouping criteria, will result in perfectly positive situations. Variables, such
as outside-of-class interactions or a student’s temperament on the day o f her interview.
may result in an emotional response that is not entirely reflective o f the classroom
situation. Therefore, only overtly positive or extremely negative responses were
considered to be significant.
A success index o f a group was determined by summing the judged values (+ - or
0) o f the responses given by each group member. Any group whose combined responses
led to non-negative outcome was deemed successful, with the exception o f four-member
groups where the responses were split evenly between positive and negative. It was
decided that a group where at least half o f its members were unhappy was unsuccessful.
The results appear in Table 4:
SECTIO N ONE
Round One
+ ,+
total=+2
+,-,o,o
total=0
+ ,+ ,0 , 0
total=+2
-- -
-. 0,0
total=0
to ta l= - 3
I o ta l--I
+ ,+ ,+ ,0
total=+3
+ ,+ ,0
total=+2
+ ,+ ,+
total=+3
W ta l= - I
+,-,o,o
Round Two
Round Three
+,-,o
+ ,+ ,0
total=+2
Iotal=O
+ ,+ ,+ ,+
total=+4
-A O
SECTIO N TWO
Round One
Round Two
Round Three
- .0 ,#
W ta l= -I
+ .0 ,0
tota l= + 1
+ ,+ ,+ ,+
total=+ 4
+ ,+ ,+ ,+
total=+4
-,-,o
W ta l= - I
+ , + ,0
total=+2
to ta l= - 2
+ ,+ ,+ ,+
total=+ 4
+ ,+ ,+ ,+
total=+4
+ ,+ ,0
total=+2
+ ,+ ,0
to ta l= + !
+ ,+ ,+
total=+3
+ ,+ ,+ ,+
total=+ 4
W ta l= O (n eg )
+ ,+ ,+ ,+
total=+ 4
-A O
+, + ,-,-
TABLE 4: Success indexes of the 28 groups considered in the follow-up. Each cell
shows the success index as determined by student responses. The response o f each
student is symbolized by a +, -, or 0. Success indexes were computed by summing student
responses within a group. Groups with a non-negative outcome were considered
successful. Groups with a negative outcome (shown in italics) were considered
unsuccessful. The symbol 0 is used to denote individuals who did not participate in
interviews.
54
O f the 28 groups studied, 21 (75%) demonstrated characteristics o f successful
groups when the selection criteria was applied. This is a considerable increase compared
to 5 o f the 14 groups (36%) discussed in Chapter II.
Demonstrations o f Improved Group Dynamics
It retrospect, w e determined that the following major problems contributed to
group dysfunction in the Spring 1998 class:
1. Extreme heterogeneity
2. Diversity in group pace
3. Presence o f a distinct leader (lack o f general student participation)
The new and deliberate grouping strategy used for the Fall 1998 students was able to
greatly reduce the group dynamics problems caused by each o f these factors.
Extreme Heterogeneity
It appears that actively segregating the previous-knowledgeable component o f the
class virtually eliminated tension due to perceived differences in scientific background.
Both previously-knowledgeable and non-previously-knowledgeable students were
delighted with being able to “connect so well” with other group members. Nonpreviously-knowledgeable students seemed relieved that they were working with people
who possessed an equally weak physics background. In fact, grouping students with
similar deficiencies in scientific background seemed to enhance the course for some
students:
We worked pretty well. We were all basically on the same kind o f level o f
understanding, and so we all, like, when we didn 't understand something, we all
didn 't understand it, which was nice, because then we could work through it
together. Or i f one person got it, then they could help the other...
Cooperative-uncertain student
I think we worked really well together. With the circuits section, it was really fun.
A nd we were all kind o f learning together, ’cause none o f us had a clue about
circuits.
Cooperative-confident student
It made such an impact on me. Iw as really afraid o f physics. I didn 't have any
physics in high school, and I'm thirteen years out o f high school and haven’t had
sciences since then, and I was really nervous on how I would understand or react
to how this class was being taught. But with them there, it was wonderful.
Cooperative-confident student
Similarly, the previously-knowledgeable students expressed enjoyment at working
together. The interactions reported by the Fall 1998 previously-knowledgeable students
w ere strikingly different from those in the Spring 1998 class. There are many more
descriptions o f general group discussion, and, unlike the Spring 1998 component, these
previously -knowledgeable students describe experiences in which they learned from their
fellow groupmates:
INTERVIEWER: What did the interaction look like?- Was it almost all the time,
or did you kind o f work on your own and then ju st come together as you needed
to?
STUDENT: No. Itw as all the time. I don’t think we ever did it by ourselves.
Maybe a little bit, y o u ’d help somebody with it as went along, but never by
ourselves and then come together and discuss it. W e’d discuss it as we went.
Previously-knowledgeable student
56
STUDENT: We were really, really... you know, argued about - not argued, you
know, in the negative sense - took things apart, took concepts apart, and we
really picked at ’em.
INTERVIEWER: So it was a little more analytic?
STUDENT: Mm-hmm.
Previously-knowledgeable student
There also is evidence to suggest that a segregated cooperative group learning style
proved more beneficial to the learning o f the previously-knowledgeable students. Unlike
the students in the Spring 1998 class, who saw themselves more as helpers or teachers
than learners, and who professed to not gaining a great deal o f new understanding, some
o f the Fall 1998 previously-knowledgeable students reported that their understanding o f
physics increased dramatically. A comparison with a response from a previouslyknowledgeable student from the Spring 1998 semester suggests just how much the
situation o f previously-knowledgeables improved:
Like, physics in high school, I don’t remember a lot o f it, because it was more,
“Here are the formulas you memorize, ” and you memorize what, when, and
where, and then you got the answer, but I couldn ’t tell you the formulas now. But
this stuff, I remember how we did it, and how you got there, so I could explain it.
I mean, it forces you to be able to explain the concept to someone who knows
nothing about it. A nd us, it was like we had to work backwards; we knew the
concept, but we had to get back to where it actually began, and work up to that.
So that was good.
Previously-knowledgeable student
Fall 1998
[My understanding o f circuits] is just inherent, I guess. I ’ve seen it done so many
times, I can probably do it almost in my sleep. I know how some things should
function and how they should not..., and then my other group members pretty
much didn’t have a clue.
Previously-knowledgeable student .
Spring 1998
D ue to minor problems with the initial grouping instrument***, one group did end
up with a situation o f extreme heterogeneity, and as with the situations observed in the
Spring 1998 class, this group turned out to be unsuccessful. In this case, a cooperativeuncertain student had taken an introductory astronomy course, and thus functioned as a
previously-knowledgeable student during a section studying the phases o f the moon. As a
fellow group member recounted, the previously-knowledgeable student’s experience with
the subject was more confusing than helpful, and ultimately this groupmate was forced to
seek outside assistance in understanding the topic:
INTERVIEWER: D id [IT ’s astronomy experience] help or hurt to kind o f give
you hints as to how it [the phases o f the moon] worked, or you wish that you had
been able to do it without that input?
STUDENT: You know, I thought it was going to help, but I never really
understood, and I know she tried, and she was real patient... I never,really quite
understood.
INTERVIEWER: What about now? Do you feel like you have a model..:?
STUDENT: Yeah. Now I Ve had a different person explain it to me, and that
made all the difference in the world.
This example serves to further substantiate the need to separate students with different
levels o f experience.
Diversity in Group Pace
Reported group pace problems were much less common among the groups in the
Fall 1998 class. Several students commented that they appreciated not being rushed, and
the comfort with the group pace seemed to relate closely to the level o f understanding
students believed they had achieved:
*** One student failed to identify that they possessed previous experience in astronomy. A solution to such
a problem would be to specify course topics on the grouping instrument, so that students can be considered
This group actually seems to be the best group I ’ve had, because there’s nobody
that has one particular role, and it occurs to me that everybody, like they ’re with
each other - we d o n 't have one person that we ’re trying to bring up to speed and
then we do n 't have one person that’s trying to push everybody to go further - y o u
know, we all work well together and we all work at the same pace and all seem to
have the same level o f understanding:
INTERVIEWER: What was the best attribute o f this group? What worked really
well?
STUDENT: I "think the pace that everybody was able to set... Some days we went
faster, some days slower, but everybody was on top o f things, everybody was
interested in the topic, it seemed like.
I think that we ’re all working about the same pace. We all have pretty equal
understandings. Like we all did really well on the last test, and I think that was a
big part o f being able to work well.
O f course, not all students felt comfortable with the pace o f the group. One
cooperative-confident student explained problems she encountered with her first group:
It was very fast-paced. I always fe lt like, “Oh, I gotta keep up with everyone
else, ” and it was like I always felt, like, really rushed, and there was people in the
group [sic] tend to be really fa st with their work. Whether they got it right or
wrong, they wanted to go ahead, ’cause it was, “This is an answer to this.
Whether i t ’s right or wrong, well, we ’ll ju st move on. ” So there was a. lot o f
that...
Cooperative-confident student
However, in this particular instance, the student who complained o f a fast-pace was also
remembered by her fellow group members as quite often being tardy:
She [SS] would come late, and we 'd explain things to her several different ways,
and she still didn’t get it, and it was frustrating, ’cause the three o f us were ready
to go on, but we couldn't...
It could be that the perceived pace dichotomy had less to do with the student not being
able to w ork at the pace o f the group and more to do with the fact that her group often
selectively previously knowledgeable. This is covered in the Appendix.
59
devoted time and effort to help her catch up. That could explain the frustration on SS’s
part at generally being told the answer and not necessarily participating in the reasoning
process.
•
M ore often, pace problems presented themselves when a minority o f students
worked faster than the rest o f the group. Rather than coming from a difference in
understanding, however, many students reported that pace problems stemmed from social
interactions. In some particularly chatty groups, one or more students would pull ahead
while the other students socialized:
I fe lt that she, I d o n ’t know, ju st kind ofjum ped apart from the group, FH did,
while me and PP were trying to work as a group, and I think that’s what kind o f
lead to us ju st stopping our work and then socializing and stuff...
Because o f the fa c t that we were more social, it was harder to focus from one
experiment to the other, ’cause we 'd kind ofjum p into social things —s o l ’d kind
o f you know, WP and I, the hoy in the group, we would kind o f move ahead and
read through it, and then maybe the other girls would come along with us.
STUDENT: I know F N was frustrated with AD and I ’cause we would talk so
much, and he would ju st go on and do the work himself
INTERVIEWER: Is that right? So he ju st kind o f operated independently?
STUDENT: Yeah, it wasn't more like a group effort. It was more an independent
effort when it came down to that.
A deliberate attempt to eliminate the pace dichotomy also had an unfortunate, but
not surprising, side effect. Because students were now being grouped with people that
didn’t rush them along or slow them down, it became more difficult for some groups to
stay on task during class, especially when they were not overtly interested in the topic at
hand.
60
[SJince we knew each other so well, or we fe lt like we got along so much better,
we talked a lot more, and we got distracted a lot easier. So, like, we didn 't keep
up with the agenda so well....
Sometimes we ju st didn't want to work, so we would, like, talk about our weekends
or something like that. But we got everything done, but sometimes we just didn't
want to work.
It did n ’t work out very well, actually, as fa r as learning and stu ff like that. It was
a lot more chatty about social issues and stuff like that. So we got our work done,
you know, but it ju st didn't work out very well because we weren'tfocused. You
know, as fa r as our learning styles and everything, that was completely fine, but
socially, we decided we had more o f an aptitude fo r that area.
It is difficult to tell the extent to which these problems stem from the course versus the
groups themselves. It is possible that students became more social as the material became
too difficult, or not difficult enough. This is discussed further in the section Problems
Arising during Implementation starting on page 62.
Presence o f a Distinct Leader
The idea o f a identifiable group leader emerged far less often in Fall 1998 groups. .
Many students described their groups as having no single leader, and even that the role o f
leader seemed to rotate naturally . A lot o f interviewees remarked that a defining factor in
their favorite group had been a lack o f distinct hierarchy:
This group actually seems to be the best group I ’ve had, because there’s nobody
that has one particular role:
We all share an equal part within the group.... In this group, no one person takes
the lead.
61
STUDENT: We all work together like that... I can't explain it. I t ’s an equal...
INTERVIEWER: There ’s not really a leader? You all kind o f pitch in, and no
o n e’s fighting?
STUDENT; I would say that we all take the lead when we need to, but i t ’s not
pushy...
In the few groups where hierarchies did appear, the “leaders” seemed to be resented by
students, rather than treated as authority figures as was done with the previously
knowledgeable students in the Spring 1998 class.
He % like, go off. on his own little tangent and then come back as “Mister
Authority, ” like, this is how I did it, you know, type o f thing, which is totally
contrary to team effort. I fin d myself more and more like, “We ’re a team here. ”
Sometimes I ’m more o f a follower in. this group, but it seems like a lot o f times
T P ’s a leader, but sh e's not always right in what sh e's saying, so I have to, like,
be a follower but at the same time I have to lead and tell all the group that, no, I
d o n ’t think this is right.
In cases where “leaders” were perceived, they seemed to be more a result o f group
dysfunction than experiential differences. The two situations mentioned above appeared
to be the result o f tension between students due to genuine personality conflicts. It should
also be noted that both aforementioned situations involved non-traditional students, who
presented an unexpected source o f problems, which are discussed in the following section.
These are not, however, the only groups in which leaders attempted to emerge. In
fact, none o f these “leaders” refer to themselves as such in their interviews, so we must
consider that the perception o f unwanted “leaders” stemmed from conflict among group
members. It is easy to see how a student might interpret disagreements with a groupmate
as evidence o f that person trying to “run” the group. This conclusion is supported to
some extent by the fact that all o f the groups containing “leaders” fell into the unsuccessful
62
category.
Problems Arising during Implementation
It would be unrealistic to expect that all problems arising in cooperative group
learning situations could disappear merely by grouping according to one or two
characteristics. While segregating previous-knowledgeable students and creating groups
from adjoining categories resulted in a dramatic improvement in success rate as
determined by our criteria, other secondary problems presented themselves. While these
problems were not nearly as dramatic as those demonstrated in the Spring 1998 semester,
they appeared solvable with certain modifications to the classification instrument, and
thus w e choose to address them. Some, such as problems associated with non-traditional
students, seem to be somewhat independent o f our grouping techniques. M ost o f these
secondary problems, however, likely occurred as a result o f our grouping criteria. In this
section, w e will discuss the problems at hand and possible solutions.
Non-traditional Students
W ith the segregation o f previously-knowledgeable students, tensions due to
physics experience were generally eliminated. However, previously unobserved problems
emerged between traditional and non-traditional students. “Non-traditional” here refers to
students who have spent a significant time (several years) out o f school. M ost o f the nontraditional students considered were married, had small children, and were employed full­
time in addition to being full-time students. Some traditional students had similarly full
63
lives, but, in spite o f marriage, children and work, had never left school, and thus viewed
themselves differently than non-traditional students.
In the Spring 1998 semester, the class considered had three non-traditional
students (16% o f the class), only one o f whom was interviewed. The Fall 1998 semester
had one non-traditional student in the second section (6% o f the class) and four in the first
(25% o f the class). Fourttf o f these students were interviewed.
All o f the non-traditional students were quick to distinguish themselves as such.
They demonstrated different priorities, than their traditional counterparts, with a large part
o f their focus on their lack o f recent experience, as well as outside commitments such as
family. This was borne out both in their interviews and in the interviews o f their more
traditional group members:
INTERVIEWER: So what was the worst thing about that group?
STUDENT: We had one partner - CD - who had ... - and I d o n ’t even know i f it
would even classify as lWorst ’ - but it was tough when somebody came in late. .
She has children... You know ju st late, as fa r as the class. You know, like h a lf
hour, fifteen minutes, whatever, when everybody has started and w e ’d have to
catch somebody up.
Traditional Student
[The two non-traditional students in my group], talk about each other’s husbands,
and one o f them has a kid, and they talk about the kid, and w ho’s going to watch
the kid and stu ff like that.
Traditional Student
Well, fo r myself, a lot o f my physics experience... was twenty, twenty-two years
plus, and all three o f the other members in the group were pretty much fresh out
o f high school.
Non-traditional student
W One o f these students dropped the course halfway through the semester.
64
I was really afraid o f physics. I didn’t have any physics in high school, and I'm
thirteen years out o f high school and haven Yhad sciences since then, and I was
really nervous on how !w ould understand or react to how this class was being
taught.
Non-traditional student
I ’m so much older than a lot o f the students in the class. I don Yknow... I go back
and talk to my husband about it all the time...
Non-traditional student
All o f the interviewed non-traditional students displayed consistent opinions on
tw o subjects: their interactions with traditional students and their attitudes towards
cooperative group learning. It was clear that, in many cases, the non-traditional students
questioned the intellectual priorities o f their traditional groupmates. The non-traditional
students took their roles as future teachers very seriously, and focused strongly on the
context o f the course material as part o f their upcoming careers®. As one non-traditional
student put it:
STUDENT: [TJhe only thing that frustrated me was their youth not allowing them
to look beyond the content material in the course, and to actually look at what the
reason fo r the course was, and that would be to not necessarily pick up on
content.... [T]hat's the thing that frustrates me in general.
INTERVIEWER: So the younger students had a really hard time seeing that
perspective?
STUDENT: Yes. [A traditional student would say,] “Oh, I know this!” “Now
wait a minute. That’s not exactly what we ’re s ’p osed to be doing here. Okay?
We ’re s ’p osed to be learning about presenting it to the students, and fully
understanding the contents before we present ’em to the students. ” And a lot o f
’em just, “Oh, yeah, yeah, yeah. I know that. ” Well, you might know that, but
you need to know several different ways to present it. A nd that’s one good thing
about the groups was because there were so many different perspectives o f how to
® It could be said that their focus on course material in the context of teaching was overly strong, as
Physics by Inquiry is intended to be a content, rather than a methods, course. Regardless, the difference in
perspective proved a source o f frustration for non-traditional students.
look at what you were doing that it would give you an idea of, “Oh, hey! Well,
I ’m going to have students just like that in my classroom, so I need to know how .
to get the point across to each o f them. ” So I look at it from that perspective.
This particular non-traditional student found herself regarding group members more as
future students and less as peers. In this case, the effects were benign, and some o f her
fellow group members expressed enjoyment with her relentless pursuit o f method:
S h e’s going into her student teaching next semester, and sh e ’s got a science
emphasis... The way she thinks is ju st amazing to me. I mean, her logic can ju st
go through anything, and she can explain it to you logically so that I understand
it.
Unfortunately, the self-imposed differentiation by non-traditional students didn’t
always have such positive results. More often, the non-traditional students saw
themselves as more worldly than fellow group members. This led them to somewhat ..
devalue the intellectual contributions o f more traditional students. In one particularly
negative group interaction, a non-traditional student described her interactions with a
traditional student:
[W]e have a person who is total, like, traditional age fo r that class, and so maybe
ju st in that she was feeling defensive at “These two older students ju st think
they 're going to come in. I'm going to give my ideas... ” So it could be ju st an
age issue, you know, a maturity level, you know, and I ’m more apt to be vocal,
and she was like, defending her right to be nineteen or whatever, you know.
This comment expresses a great deal o f frustration with the traditional student, and any
contribution is disregarded as lacking maturity. The traditional student in question
described the same interactions:
66
STUDENT: S h e'd never taken it [physics] before, but she ju st took it upon
, herselfto feel that she. needed to be up with them [the preyiously-knowledgedble
students] with us in the group. She acted very singularly. It w asn’t a group with ■
her. “Come on, le t’s rush through this. L e t’s get done with this. ” So Iguess how
I acted, I tried to fig h t my way back, you know, to fin d a place, ’cause CD and I
would be trying to understand concepts together, and AD would ju st be like,
“Come on, ladies, ” and then sh e'd be telling us about her weekend, you know,
one second, and the next second sh e'd be rushing through.
INTERVIEWER: So how would you pull her back?
STUDENT: How would I pull her back? I 'd say, “Listen, we ’re not even done
with this problem, you know? ” She'd be two ahead o f us. So sh e ’d close her
mouth, you know, i f she was frustrated with us, basically, close her mouth andjust
start working on her own, not help us, not work together as a group.
It would appear that this traditional student was actively trying to contribute, but this nontraditional student had assumed a role similar to one seen in a previously-knowledgeable
student from the Spring 1998 semester. In this case, traditional vs. non-traditional status
served to promote a hierarchy within the group. O f the 28 groups, three suffered from
this problem, almost half o f the unsuccessful groupings. It would appear the life
experience can create group tension in much the same way that physics experience does.
Fortunately, there were also groups containing non-traditional students that were
successful. These groupings seemed to follow specific patterns. If non-traditional
students were isolated within groups (6 groups), or if the entire group was composed o f
nori-traditional students (I group), the group was successful558. However, if a group had
non-traditional students in the majority with traditional students in the minority (2 groups)
- in these cases, tw o non-traditional students with one non-traditional student - the
m One case with a single non-traditional student was unsuccessful. However, the lack o f success in this
group seemed related to group coherence rather than the role o f the non-traditional student; one group
member virtually stopped coming to class and generated a great deal of animosity with her fellow group
members.
67
groups were unsuccessful****. In both the unsuccessful cases, the traditional students
expressed frustration at the lack o f group functionality:
We had a lot o f animosity. Like I don't even talk to AD now. It sounds so petty,
but it was so frustrating to learn with her. I could not learn with her. I didn 7
learn anything.
This group, I might as well ju st be working on my own on this one. I 'd have a lot
less headaches.
It is interesting to note that the one group which consisted o f all non-traditional
students contained a student who was continually behind, had trouble with the course
material, and was one o f the two students who ultimately dropped the course. In this
case, though, one o f the same non-traditional students who expressed frustration at a
traditional group member in a previous quote described a far more positive experience:
VN, I hadjust met then, that was the first time I had ever met him before, and I
thought he was super in the group. You know, it didn 7fe e l confrontational.
Some people are very confrontational, and I thought we interacted really well, in
terms o f like my ideas, his ideas, or me questioning something that he had. You
know, very comfortable. It didn 7fe e l like Iw as challenging him on any ideas...
It would appear that she was willing to consider a fellow non-traditional student as more
o f a peer than the younger, but more academically-capable, student she so actively
conflicted with. This strongly suggests that non-traditional student status should be a
deciding factor in our strategy o f group organization.
Another somewhat unexpected outcome o f the interviews with non-traditional
students was their attitudes towards group learning. While they each advocated it as a
good teaching style, they expressed reservations about participating in it personally:
**** One successful group, containing two non-traditional students and one previously-knowledgeable
student, was reported in the Spring 1998 semester. It is possible that physics experience can be
.
68
I'm not a big group worker anyway, so when they said at the beginning o f class
that it was going to be group work, it was ju st kind of, “Oooh... ”
I ju st go like, “Gosh, we have to be in groups, ’’ and I ’m a non-traditional student.
I could ju st do this stuff by myself You know, other group projects that we have
in our other classes too, that takes me away from my family. Well, this one
doesn't, but ju st stu ff like that. Being a non-traditional student, I don’t like,
groups, but as a teaching strategy... I can see where it would work well...
This is especially interesting when one considers that these same students each expressed
at least one very positive group experience. It would seem that the students in question
experienced more difficulties with the mechanics surrounding group w ork (commuting,
making time for group projects, etc.) than many traditional students. A solution to this
might be to take these problems into account as an instructor and try to minimize
extraneous hassles (traveling great distances to meet with groups, scheduling difficulties,
etc.) as much as possible.
. .
Previouslv-knowledgeable Students and Expected Coursework
While the segregation o f previously-knowledgeables improved their learning
experiences, grouping students with previous physics experience did lead to an unexpected
.
problem. Because students were able to start communicating at a more advanced level,
quite often group discussions were much more in-depth than the course material
anticipated. Previously-knowledgeable students often found themselves bored or
foundering:
considered equivalent to life experience in such a case. .
69
INTERVIEWER: Were there times in the curriculum when you or members o f
your group were bored or wishing you could skip ahead?
STUDENT: A ll the time.
INTERVIEWER: A ll the time? Pace fe lt too slow fo r you?
STUDENT: Yeah, it did. All o f us had had physics before, so ju st the idea that
we did it all in the tenth grade...[I]n the book, the questions are so repetitive that
there’s times, like i f yo u look at my notes, I have like six things that say, “See f o u r .
point one, ditto, ditto, ditto, ditto, " ’cause I ’m not writing it all out.
I think sometimes it would get a little long; the two hours would be long. That
unit was real, kind of, basic, I think. We 'd all done it before.
The students generally traced their frustrations back to the text. Many found it too basic
to accommodate their previous encounters with physics, and at the same time, full o f
seemingly “trick” questions:
I didn ’t really like the book, and that I think contributed a lot to the frustration in
the [second] group, because the book, i t ’s supposed to be a college-level book,
and I did n ’t think it was. I thought it was pretty basic. I think that was one o f the
major frustrations was the book, and how the book worded the questions or
worded the experiments or what to do.... You learn the concepts, and i t ’s very
basic, and that helps you, but i t ’s like, some o f the basic concepts that I did get
out o f high school I came in and I ’m like, “Well, I thought I knew this, ” and I
actually didn 't know it.... Itw as really frustrating, and I think the book really
added to it. It was really frustrating, the book.
In spite o f previously-knowledgeable students’ frustration with the basic nature o f the
course, they tended to move at the same pace as the rest o f the class and expressed
frustration with the instructor at being asked to learn material at an accelerated pace:
In the particular groups that I was in - we ’re in this advanced group - w e'd go
faster than the rest o f the class, and that really bothered me sometimes because
we 'd be given all this extra stuff to do and the rest o f the class wouldn 't do it, and
I thought, “Well, i f i t ’s supposed to be beneficial, why aren’t they doing it? ” A nd
so w e ’d fe e l frustrated with some o f the extra stuff we had to do.
The seeming contradiction o f boredom and resentment towards an accelerated pace was
difficult to reconcile. In one case, a previously-knowledgeable student in Section Two
became so frustrated with the banality o f course material that she stopped coming to class,
much to the frustration o f her group:
STUDENT: A T ’s attitude really bothers me, because, like, she doesn 't care. She
thinks i t ’s way to basic fo r her, so she ju st doesn’t care. She doesn’t show up to
class. I mean, she comes to class, and i t 's like, she ’ll go through it at home, and
she doesn’t expect us to catch her up because she already knows it, but I don’t
know. She doesn’t like the class, she doesn’t like [the instructor], and she doesn 7
want to be there and it shows.
INTERVIEW ER: That unfortunately is hurting you, I guess.
STUDENT: Right, yeah. I t ’s like, “Well, y o u ’re apart o f our group, and our
group depends on all o f its members to work together. ” A nd lots o f days sh e’s
never even there. Andthen when she is there, sh e’s got a negative attitude about
it.
What frustrates me is ... refusal to see beyond what was in front of, and that goes
back to A T ’s, “Oh, Iju s t don 7 see why we have to do this. This is ju st so stupid. ” .
Well, yeah, you may think i t ’s stupid, but hey, to get out o f this class,, this is what
w e ’ve got todo.... I mean, we ’re going to make it out all right, but i t ’s ju st
frustrating to me to have to sit and listen to people like that, but I have to deal
with it, because I know there's going to be students like that in my classroom.
Unfortunately, in the process o f complaining about this particular student, her fellow
groupmates reveal their own similar frustrations with the course and the material.
It would appear that, by improving group dynamics for the previouslyknowledgeable students, we have, in the process, diminished the effectiveness o f the
course for them. Group learning strategies appear to work best when the group members
are challenged by the subject; that failed to happen here. This brings into question the
motivation for previously-knowledgeable students to take a course so specifically designed
for non-majors. Fortunately, one student revealed her motivation for being in Physics by
Inquiry even as she expressed her disappointment:
It ju st seemed like, fo r a college class, I ’m learning more how to teach physics. It
didn't do a very good joh o f that. That's I guess what I thought the class was a b o u t-y o u know, how to teach the basic ideas o f circuits and math and things
like that - and it ju st fe lt like we were doing the work, but we weren’t really
learning how to teach it better.
Apparently, at least some o f the previously-knowledgeable students approached this class
as a methods, rather than a genuine physics, course. Given the negative reactions o f the
students to the content, Physics by Inquiry is inadequate on both counts.
These data suggest that improved grouping methods could have a drastic impact
on curriculum structure. In this case, improved group dynamics led to frustration with the
course material that, had students not been organized in this manner, probably would have
manifested itself as frustration with other group members. Since going back to previous
grouping strategies would contradict the findings o f this study, the best remedy for such
problems may lie in fundamental curricular changes aimed specifically at previouslyknowledgeable students.
Group-switching Backlash
Far and away, the initial student reaction to group-switching was negative. Some
students eventually enjoyed the groups they switched into, finding that a change o f pace
helped their learning:
I disliked it at first, after switchingfrom the first to the second, because we were
ju st starting to click. You could pretty much count on the first group, at least a.
few o f the people in our group, being there everyday on time, and so you started
to notice everybody’s routines and kind o f how they thought and whether you 'd
fe e l like a total dink i f you asked a question o f somebody else or asked somebody
to help you out, and everybody was really helpful in that way. A nd then once we
got going along the switching groups, it was great to get to know other people
and be able to talk to other people.
For the most part, though, students did not enjoy their personal experiences. As o n e .......
student commented:
M ost o f the people I Ve talked to in the class, their first group was their favorite
group. You know? I mean, MD ’s first group was her favorite group, and R B ’s
first group was her favorite group.
Even for students who cited their second or third groups as their favorites, it was apparent
that change was often more difficult than rewarding:
I d o n 't like it. I d o n ’t know... You get used to a group and you like staying in and
finishing up, you know, ’cause you work so well together. I can understand
somebody should maybe go talk to the professor or the TA [teaching assistant] or
whatever and tell them that they 're having a problem with their group, but we
weren't having any problems, and PP and I were both begging [the instructor] to
stay in our group the first time.
In general, Ife e l that i f we [the first group] would’ve ju st kept together, and not
had to have split up, that would’ve been the best.
I
Ify o u think you d o n ’t understand something, you ’re afraid to ask ’cause i t ’s a
new group. You d o n ’t have the comfortableness [sic] yet.
I d o n ’t like, it ’cause you have to get into a new group and... i t ’s a whole new
style, kind o f like, you know.
I think that i f you stay with the same group fo r a long time, you get so
comfortable that it would be hard to work with other people. You just get so used
to the role that you ’re in - 1 don’t know i f that makes sense - like, i f you ’re the
one that’s always speaking up fo r your group, you might get so used to that it
would be hard to work with another group...
I think i t ’s really hard when you get thrown in with a bunch o f people to try to get
a foundation, and you finally get one - like in that first group, we started working
well together, and started enjoying it ’cause we moved along quickly and I
started, you know, understanding how it all worked, and it really jostled me, I
. think, to switch into a new group. Then you Ve got to start all over again, and I
think that was a kind o f struggle fo r me.
73
The preference o f dynamics within the first group is reflected by the success rates o f
groups at each round o f changes. For the first round, nine out o f ten groups were
successful (90%)tttt, compared to five out o f nine (56%) for the second. The high
success rate o f the first round strongly suggests that the group-learning techniques
students adopted were highly dependent on their initial group interactions. Many students
described positive interactions with members o f their first group, followed by subsequent
problems within the second. This was often attributed to differences in personalities or
learning styles. Just a small sample o f the comparisons follow:
I d o n ’t know i f I would have changed anything. It seemed that was by fa r the
most productive fo r me, as fa r as what I was learning, and everything seemed to
go pretty smoothly.
WP describing first group
STUDENT: I d o n ’t know what it was, but our group didn ’t “click” maybe as well
as say the first one did.
INTERVIEWER: D idn’t click as well with the science, or didn’t click as well with
the social?
STUDENT: No. Social things were fine, and the science seemed to... we were
ju st at a different pace maybe. People were learning differently...
WP describing second group
*
*
*
tttt jt should be noted that the one unsuccessful first round group may have been due to factors beyond the
instructor’s control. Apparently, two students were grouped together who had conflicts previous and
external to the class:
STUDENT: I w o u ld ’ve lik ed to have a different p erso n besides EK. I think th a t that m aybe
w o u ld ’ve m ixed the grou p better, an d E K w a s in eve ry single one o f m y c la sses also, p lu s the
p h y sic s an d p lu s w e w ere in the sam e group together, so that w as k in d of...
INTERVIWER: Too much E K ?
STUDENT: Too much E K
In the future, such problems could be avoided by including a question in the instrument which asks
students i f their are class members they would prefer not to work with.
That group worked really well together, so I wouldn 't really say there was a
leader I think everybody pulled their own weight and did their- own.fair share,
and I d o n 't think there was really a leader or a follower.
DB describing first group
Probably the second group was [the least favorite.] Just personality... learning
styles were different.
DB describing second group
*
*
*
We all kind helped each other. We worked really well together. Like, we all were
the same kind o f people. We took a little longer to get it, but once we got it...
You know, and i f somebody else in the group didn % then we 'd keep changing the
ways o f explaining it to make sure that all four o f us understood it.
PP describing first group
They both completely understood what we were doing, and I was going a little
slower, but it wasn 't slowing them down at all. They ju st kept going ahead o f me,
and I was ju st finding myself falling behind and I was so uncomfortable with both
o f them. Like, our personalities didn’t seem to click.
PP describing second group
The students, attribute most negative interactions to personality conflicts, but a
more plausible explanation might be that the students each developed distinct group
learning styles in their initial groups. There was little disharmony in the first round simply
because most students may have had little or no experience with intensive collaborative
group learning, and thus were much more willing to compromise their learning styles in an
effort to establish individual group equilibria. After switching groups, students probably
expected to continue to interact in a manner similar to that o f the first group. However, if
group dynamics were tailored within each initial group (as appears to be the case),
switching groups would instead lead to conflict as students were forced to quickly adjust
to new learning styles. It would seem that this researcher’s assumptions about the
innocuousness o f switching groups proved incorrect, and successful grouping may only be
compromised by group switching.
There was, however, one circumstance in which group switching proved ideal for
students. This happened in situations where initially timid students (mostly cooperativeuncertains) demonstrated increased confidence as they moved to each subsequent group.
Some examples include:
INTERVIEWER: How did you behave in this group?
STUDENT: I d o h ’t know. Kind o f shy,, I guess.
DU describing first group
I opened up a little more. Like I talked more in this group, I think, ju st 'cause I
got more comfortable with physics, 'cause this is my first class, so I think I ju st
got more comfortable with everything.
DU describing second group
Again, [ I ’m] more outgoing, ’cause I understand much better, and I ’m getting
more comfortable with it.
DU describing third group
*
*
*
I did a lot o f observing, I would say. I didn 'tfeel that I could be a fu ll
participant, I guess.
ML describing first group
I was more interactive, I ’d say, with that group...
ML describing second group
I ’m actually interactive. I help With the problems. I fe e l like I ’m doing this
instead of, you know, kind o f doing it and then having take it home andfiguring it
out. You know, I'm doing it in class and I ’m figuring it out.
ML describing third group
In these situations, group switching proved beneficial for the cooperative-uncertain
students. W e speculate that their increase in participation and confidence in the subject
would not have taken place had there not been group switching. These students were in
the minority but still a noticeable percentage (6 out o f 34 students) o f the two sections.
W e are left with the dilemma o f which scenario should take precedence: should group
switching be avoided so that most students can remain in harmonious groups, or should
w e continue with group switching to promote increased confidence in science
among a minority o f the students?
Improvements for Future Implementation
Based on the new problems encountered as part o f the implementation o f categorized
grouping, this researcher recommends the following three guiding principles:
0
Deal with non-traditional students in much the same way as previouslyknowledgeable students.
Non-traditional students seem to function best in two situations: grouped exclusively
with other non-traditional students or isolated entirely from other non-traditional students.
I f a t all possible, groupings with other non-traditionals seems favorable. This allows for a
more uniform frame o f reference within the group, and students might find it easier to
accommodate each other’s schedules for meetings outside o f class. However, this
grouping o f non-traditional students should be secondary, after previously-knowledgeable
students have been segregated.
0
Provide a more challenging environment for previously-knowledgeable students.
The problem o f fully engaging previously-knowledgeable students as active learners is
slightly more delicate in the context o f this course. We must accommodate the fact that
previously-knowledgeables are not being challenged by the traditional course material, but
feel that it is inappropriate to be presented with different expectations than the rest o f their
classmates. To avoid the in-class dichotomy altogether, it might be best that previouslyknowledgeable students not be included in this particular inquiry-based physics course at
all, and may benefit from a more advanced traditional course with weekly inquiry-based
recitation sections®*. This would not only present the previously-knowledgeables with
more challenging material, but it would avoid the expectation that Physics by Inquiry
should function as a teaching methods course.
0
R econsider th e value o f group-sw itching.
The usefulness o f group-switching is dependent on the expectations o f the instructor.
I f group-switching does not occur, it seems more likely that categorically grouped
students will end up in harmonious groups. Unfortunately, withdrawn students (such as
cooperative-uncertains and passive-uncertains) may not develop the same level o f selfconfidence that they might if group switching occurs. In the context o f a course such as
Physics by Inquiry, which is geared towards future teachers, the benefits o f group
switching may outweigh the drawbacks. If these students are eventually going to present
science to their own pupils as knowledgeable resources, it is important that they all
develop confidence in themselves as scientific knowers and learners. I f groups are not
switched, the more withdrawn students may never reach this point.
# # M n ntana State University does provide two semesters o f such a physics course, which includes three
hours o f lectures with active discussion and demonstrations, and a two-hour inquiry-based recitation
section similar to those used in Physics 2 0 1, but presenting material at a more advanced level.
W ith these modifications to our model o f cooperative group organization, we are
hopeful that group dynamics success rates may be improved even further. We will discuss
the implications and possible uses o f this more complete model in Chapter 5 '.
/
79
CHAPTER 5
DISCUSSION
Conclusions o f our Study
This research project was initiated to better understand the nature o f group
dynamics in small cooperative learning groups in the context o f an inquiry-based
introductory physics course for pre-service elementary education majors. The ultimate
goal was to determine the most relevant factors in creating groups where all students
participated actively and enthusiastically in the learning process. The results o f this study
strongly suggest that positive group interactions for cooperative learning situations are not
only possible, but can be predetermined with reasonable accuracy for this particular
population.
The primary predictor o f group success is relative previous experiences with the
subject matter within the group. Students with similar backgrounds in conceptual physics
worked together most effectively, and students with disparate backgrounds (a high school
physics course versus no high school physics training) tended to experience conflict and
animosity towards each other, often to the detriment o f learning by all involved parties.
Within groups o f students with similar physics backgrounds, a secondary predictor
could be derived from students’ self-confidence levels. Students could be classified into
four categories:
80
o
aggressive-confident,
o
cooperative-confident,
o
cooperative-uncertain,
o
and passive uncertain,
which reflected their levels of confidence as physics learners. It was found that extremely
outgoing and self-confident students (aggressive-confidents) did not interact successfully
with students who needed a great deal o f patience and encouragement from their fellow
groupmates (passive-uncertains). The most successful groups also possessed students in
consecutive categories, so that, for example, a group containing a cooperative-confident, a
cooperative-uncertain, and a passive-uncertain would be more successful than a group
consisting o f an aggressive-confident, a cooperative-uncertain, and a passive-uncertain.
W hen all o f these factors are taken into account, our results show that successful
i
groups can be orchestrated. Our original categorization instrument, which took into
account previous physics experience and student self-confidence, enabled us to organize
groups quickly and efficiently, and also produced a dramatically higher group success rate
than observed in the Spring 1998 semester.
However, data from the Spring 1998 implementation also brought to light
weaknesses in our model. We found that a student’s “life experiences’’ produced results
similar to, albeit less noticeable than, previous subject experience. Students with nontraditional status occasionally conflicted with more traditional students. However, this
only occurred when traditional and non-traditional students were grouped together such
that the number o f non-traditional students in the group exceeded the number o f
81
traditional students. In situations where there was only one non-traditional student in a
group, the differences in “life experience” did not seem to impact the success o f the group.
O ur original model did not take the “life experience” factor into account, but
modifications to the instrument, such as questions concerning students’ age, marital status,
and self-classification as traditional or non-traditional status might allow us to create even
higher group success rates when this technique is used in the future.
W ith factors associated with unsuccessful grouping better controlled, two
problems with the nature o f the Physics by Inquiry course structure became apparent.
First, w e found that when students with strong physics backgrounds (previouslyknowledgeables) were grouped together, their group dynamics improved, but their interest
in and satisfaction with the course decreased. They expressed boredom and frustration
with what they saw as banality and repetitiveness within the course material. Furthermore,
their main motivations for taking the course seemed to. be an interest in the teaching
methods rather than the subject material. Since Physics by Inquiry is designed to be a
introductory content course for non-majors using cooperative group learning methods,
rather than a teaching methods curriculum using inquiry-based physics as a vehicle, we
believe that current course design is insufficient to further educate previouslyknowledgeable students, and when the categorization instrument divines such students,
they should be advised to take a higher-level physics course, and possibly automatically
exempt from enrollment in Physics by Inquiry*
* As these students are primarily education majors, any change in curricular expectations must be
discussed with and approved by Montana State University’s School o f Education. It is also possible that
the problem o f previously-knowledgeable students enrolling in P hysics b y Inquiry only exists because they
The second curricular problem stemmed from the implementation o f group
switching, first implemented in the Spring 1998 semester. Most students reacted
negatively to group switching, feeling that their most comfortable and successful group
experiences occurred in their first groupings. We believe that such strong first-group
success may be attributed to the newness o f the subject matter and the environment, and
an inherent willingness to compromise learning styles in a effort to w ork together more
effectively. However, by the time students first switched groups, they had become
accustomed to the learning styles o f their initial groupings, and found learning to function
under new circumstances extremely stressful.
Such results suggest that, unless there is a specific motivation for it, group
switching will for the most part be detrimental to students’ group experiences. It should
be noted, however, that we did observe a small proportion o f students that appeared to
benefit from group switching. In these cases, students who had originally lacked selfconfidence as learners (here, cooperative-uncertains) gained confidence in their abilities
w ith each group change. Many described themselves as active discussion and experiment
instigators by the time they reached their third groups. In a course such as Physics by
Inquiry, where virtually all o f the students are training to be elementary school teachers,
confidence in the course material and a student’s ability to work within it are Seen as vital
components o f the learning process. Therefore, it is likely that future instructors o f
Physics by Inquiry will prefer to sacrifice increased group success rates to provide a
crucial opportunity for students to gain self-confidence as learners.
are being advised to take it as a methods course. If this is the case, appropriate persons should be
83
Implications for Use in Other Environments
Our study was limited to a single, narrowly-defined setting: a non-lecture,
introductory, inquiry-based physics course for pre-service elementary school teachers. In
our comparisons with other examples o f group learning, we primarily compared studies
using other physics or physics-related courses, with occasional references to studies
examining group dynamics. However, we believe that our results suggest that our group
organization strategies are more widely applicable.
It must be remembered that Physics by Inquiry is geared tow ards non-physics
majors with extremely limited technical or scientific experience. The course is explicitly
conceptual, and the group learning environment focuses on discussion and active learning
rather than computational problem-solving. The instructors’ main goals are to introduce
students to hypothesis testing and the scientific method, and use these tools to build
rigorous, self-consistent models. There is nothing about the course structure that is
inherently subject-specific. It is also possible that courses using groups under less
interactively intense conditions, such as in a problem-solving or creative discussion
environment, can increase the likelihood o f successful group dynamics using our grouping
techniques.
Due to the highly interactive and self-regulating nature O f the observed successful
groups, we speculate that our group organization techniques would be ideal for largeenrollment courses (50 students or more) which use cooperative grouping as a teaching
tool from time to time, but where instructors are not in a position to interact more than
informed that there has been a miscommunication as to the fundamental purpose o f the course.
84
briefly with any one group. In such situations, instructors generally organize students into
groups to generate discussion and encourage active learning. I f the success rates observed
in the Spring 1998 semester are typical, it is possible that many, even a majority of,
students have difficulty functioning within their groups. While group role assignments are
used by some instructors, these may be difficult to enforce in a large-enrollment setting.
Even if students could function successfully within their roles, such a teaching tool may
prove limiting to discussion, since students will be locked into a given role throughout the
assigned period, while a group functioning with good dynamics may, and according to our
research, probably will naturally rotate roles as the discussion progresses.
I
W e hope that our results and conclusions have proved sufficiently compelling that
instructors opt to use the grouping techniques described here to facilitate their future
employment o f cooperative group learning. Successful application o f this strategy will
require the instructor designing an instrument - to be administered before groups are
organized - that answers the following questions:
1. W h a t is th e stu d en t’s background in th e subject m a tte r? The answer to
this will most likely be dichotomous (the student has or has not previously
taken a course in the subject), but it can go into as much detail as the instructor
deems necessary. For instance, the instructor could ask about individual topics
within the course if the scope is sufficiently broad.
2. Is th e stu d en t o f tradition al o r n on-traditional status? This characteristic
is easily covered with inquiries as to the student’s age, marital status, and self­
definition as traditional or non-traditional. In addition, inquiries into a
85
student’s place o f residence may facilitate group learning if the instructor wants
to encourage groups to work together outside o f class; it may be easier for
students to meet if they live near each other, and would especially ease the
burden for students who have long commutes.
3. H ow does the student function in new situations Or w ith groups of
strangers? How com fortable does th e student feel initially ab o u t the
subject m aterial? Questions like these are designed to categorize the
student’s self-confidence as a learner. M ore than the other inquiries, these may
be delicate and should probably be tailored to the course environment. This
researcher’s own first attempt at an instrument was somewhat less successful
than hoped in these areas. This is discussed at length in the Appendix.
In addition to the three topics covered above, it may be useful to take a fourth into
consideration. In our study, gender was not considered an issue, primarily because the
students w ere almost entirely female. However, in mixed gender situations, Skala, Slater,
& Adams (1999) found that in groups with uneven compositions o f men and women (two
or three o f one gender grouped with one o f another), at least one female student tended to
dissociate from the group. On the other hand, groups with even gender compositions
(two men and two women or single-gender) experienced far greater instances in which all
members o f the group interacted.' The problems experienced by female students in groups
o f uneven composition are. borne out in focus group interviews concerning students’
group experiences.23 Such results would suggest that gender should also be taken into
23 Skala, Slater, and Adams, 7.
86
consideration when groups are organized, especially in situations where an instructor is
not in a position to closely monitor group activity. The most conservative approach
would be to place students exclusively in single-gender groups; this would guard against
group imbalances caused by student absences. However, it is difficult to predict how
students would react to being assigned to such groups. For instance, might female
students feel “singled out?” This aspect o f group organization requires further study.
Future Study
This research is only our first attempt at the understanding o f group dynamics.
Only with application o f our techniques in a larger range o f teaching environments can we
see if our conclusions are generalizable. We are curious as to how widely applicable these
grouping strategies truly are. For instance, they seem to work with older students and
adults, but would they w ork with children? Are they even necessary or even appropriate
among the sufficiently young? We have also derived all o f our data from a course taught
at a typical American college, as have all o f the physics and physics-related studies we
referenced. Would our techniques work in group learning situations in a sufficiently
different cultural environment? Montana State University is associated with several
Native American tribal colleges, some o f which have offered the Physics by Inquiry
course. Anecdotal evidence suggests that instructors have consistently had extreme
difficulty interacting with these Native American students, and that active group learning
has been difficult to motivate. These reports, as well as the interview provided by the
Native American student in the Spring 1998 semester, have made this researcher curious
as to the impact o f culture on cooperative group experiences.
BIBLIOGRAPHY
Andre, K. M. “Cooperative Learning: An Inside Story.” The Physics Teacher 37
(September 1999): 356-358.
Cohen, E. G. Designing Groupwork: Strategies fo r the Heterogeneous Classroom. New
York: Teachers College Press, 1972.
'
Cottle, P D. and S E. Lunfbrd. “Increasing Student Participation in Recitation.” The
Physics Teacher 33 (January 1995): 23-25.
Emkey, W. L “A Small Group Approach to Introductory Physics.” American Journal o f
PhysicsAl (August 1979): 695-697.
Good, T. L. and S. Marshall. “Do Students Learn More in Heterogeneous or
Homogeneous Groups?” InThe Social Context o f Instruction: Group ,
Organization and Group Processes, eds. P. L. Peterson, L. C. Wilkinson, and M.
Hallinan, 15-38. Orlando: Academic Press, Inc., 1984.
Heller, P. and M. Hollabaugh. “Teaching Problem Solving through Cooperative
Grouping. Part 2: Designing Problems and Structuring Groups.” American
Journal o f Physics 60 (July 1992): 637-644.
Heller, P., R. Keith, and S. Anderson. “Teaching Problem Solving through Cooperative
Grouping. Part I : Group versus Individual Problem Solving.” American Journal
o f Physics 60 (July 1992): 627-636.
Johnson, D. W., R. T. Johnson, E. J. Holubec, and P Roy. Circles o f Learning:
Cooperation in the Classroom. Edina: Interaction, 1984.
Johnson, D. W., R. T. Johnson, and K. A. Smith. Active Learning: Cooperation in the
College Classroom. Edina: Interaction, 1992.
Laws, P. W. Workshop PhysicsActivity Guide. New York: John Wiley & Sons, Inc.,
1997.
Lunetta, V. N. “Cooperative Learning in Science, Mathematics, and Computer Problem
Solving.” In Toward a Scientific Practice o f Scientific Education, ed. M.
Gardner, 235-249. Hillsdale: Lawrence Erlbaum Associates, Inc., 1990.
Mausner, B. “The Effect o f One Partner’s Success in a Relevant Task on the Interaction
o f Observer Pairs.” Journal o f Abnormal and Social Psychology A9 (1954): 557-
90
M cDermott, L. C Physics by Inquiry: Volumes I and II. New York: John Wiley & Sons,
Inc., 1996.
■•
■
Miller, J. E., J. Trimbur, and J. M. Wilkes. “Group Dynamics: Understanding Group
Success and Failure in Collaborative Learning.” New Directions fo r Teaching and
Learning 59 (Fall 1994): 33-44.
Miller, J. E.; J. Wilkes, and R. D. Cheetham. “Tradeoffs in Student Satisfaction: Is the
‘Perfect’ Course an Illusion?” Journal on Excellence in College Teaching 4
(1993): 27-47.
Peterson, P L., L. C. Wilkinson, F. Spinelli, and S. R. Swing. “Merging the ProcessProduct and the Sociolinguistic Paradigms: Research on Small-Group Processes.”
In The Social Context o f Instruction: Group Organization and Group Processes,
eds. P. L. Peterson, L. C. Wilkinson, and M. Hallinan, 125-152. Orlando:
Academic Press, Inc., 1984.
Rau, W. and B. S. Heyl. “Humanizing the College Classroom: Collaborative Learning and
Social Organization among Students.” Teaching Sociology 18 (April 1990): 141155.
Reif, F. “ Scientific Approaches to Science Education.” Physics Today (November 1998):
48-54.
Shaffer, P. S. and L. C. McDermott. “Research as a Guide for Curriculum Development:
An Eixample from Introductory Electricity. Part II: Design o f Instructional
Strategies.” American Journal o f Physics 60 (November 1992): 1003-1013.
Skala, C , T. F. Slater, and J. P. Adams. “Qualitative Analysis o f Collaborative Learning
Groups in Large-Enrollment Introductory Astronomy.” Proceedings o f the
Astronomical Society o f Australia, in press.
Slavin3R. E. “When Does Cooperative Learning Increase Student Achievement?”
PsychologicalBulletm 94 (1983): 429-445.
Slavin, R E . “Synthesis o f Research on Cooperative Learning.” Educational Leadership
(February 1991): 71-82.
Steiner, I D. Group Processes and Productivity. New York: Academic Press, 1972.
Watson, W., L.K. Michaelsen, and W. Sharp. “Member Competence, Group Interaction,
and Group Decision Making: A Longitudinal Study.” Journal o f Applied .
Psychology 16 (1991): 803-809.
91
Webb, N. M. •“Student Interaction.and Learning in Small Groups.” Review o f
Educational Research 52 (Fall 1982): 421-445.
Webb, N. M. “Student Interaction and Learning in Small Groups: A Research Summary.”
In Learning to Cooperate, Cooperating to Learn, ed. R. Slavin, 147-192. New
York: Plenum Press, 1985.
Webb, N. M. “Testing a Theoretical Model o f Student Interaction and Learning in Small
Groups.” InInteraction in Cooperative Groups, ed. R. Heftz-Lazarowitz, 102119. New York: Cambridge University Press, 1992.
Webb, N. M. and C. M. Kenderski. “Student Interaction and Learning in Small-Group
and Whole-Class Settings.” In The Social Context o f Instruction: Group
Organizationand Group Processes, eds. P. L. Peterson, L. C Wilkinson, and M.
Hallinan, 153-170. Orlando: Academic Press, Inc., 1984.
92
APPENDIX
DISCUSSION OF THE STUDENT. CATEGORIZATION INSTRUMENT
Instrument Development
Based on the responses from the Spring 1998 interviewees, we formulated a
question tree, shown in Figure I on page 94.. The question tree consists o f a series o f ten
questions, each with a dichotomous response. Depending on the answer to a particular
question, a student progresses to one question or another, so that a student actually
answers only four or five o f the available questions. The questions focus first on
experience with physics, in order to distinguish previously-knowledgeable students.
Remaining questions deal with students’ behavior in group situations or on homework
assignments. The path traced by student responses determines his or her group learning
categorization.
For use in the Fall 1998 semester, the question tree was modified into a more
easily administered survey, shown in Figure 2 on page 95. In an attempt to emulate the
“tree” structure, any answer to a given question was accompanied with specific
instructions that students proceed to another, often non-consecutive question. The
instrument was designed so the number o f questions to which a student answered “Yes”
determined his or her categorization: four “Yes” answers constituted a previouslyknowledgeable, three an aggressive-confident, tw o a cooperative-confident, and so on.
This structure was created so that previously-knowledgeable students could be identified
immediately upon submitting their responses.
QUESTION TREE FOR PHYSICS 201
I. H a v cy o u ever
tak en P h y sic s b efore]
2B . D o y o u have any casual
e x p e r ie n c e w ith P h y s ic s?
2 A . D o e s y o u r P h y s ic s
e x p e r ie n c e s te m
(e le c tro n ic s o r a stro n o m y ,
fr o m a h ig h sc h o o l
y f o r e x a m p le )
P h y s ic s c o u r s e ? »
Jn th is
3 C . A re y o u c o n fid e n t in
3 B . H a s ) 'o u r e x p e rie n c e
3 A . D id y o u h a v e a
p o s itiv e e x p e rie n c e
y o u r a b ility to le a rn
in v o lv e d h a n d -o n
e x p e r im e n ta tio n ? V
P h y s ic s c o u rs e ?
P h y s ic s ?
4 B . D o yocl te n d l o be
4 A . D o y o u feel y o u ’ve
re ta in e d s o m e o f th e
b a s ic in fo rm a tio n
V
4 C . D f ) o u te n d to b e
o u tg o in g o r v o c al
o u tg o in g o r v o c a l
in g ro u p s itu a tio n s ?
in g ro u p situ a tim is ?
p re s e n te d in th is
. P h y s ic s c o u rs e ? X
4 '. D o y o u Ih id th a t, o n
y o u r o w n , y o u w o rk
tlir o u g h a s s ig iu n e n ts ..,
q u ic k l y
PreviouslyKnowledgeable
AggressiveConfident
a t a
X
q w e r
CooperativeConfident
CooperativeUncertain
PassiveUncertain.
FIGU RE I: Question tree created to categorize student into group learning styles. This question tree was later formulated into the
categorization instrument used in the Fall 1998 semester.
95
P h ysics by Inquiry Introductory Questionnaire
Name:_
Start with Q uestion 1 and, according to your response, proceed to the next appropriate
question; Don’t worry. You'll only end up answering four or five questions.
I- Have you ever taken Physics before?
..............
2. Did you lake a physics course in high
school?
Yes (go to 2)
No (go to 3)
Yes (go to 4)
No (go to 5)
3.
Do you h ave a ny casual experience
with Physics (astronomy, for
example?)
Yes (go to 5)
No (go to 6)
4.
Are you comfortable with fractions and
ratio m ath, and have taken algebra?
Yes (go to 7)
No (go to 8)
5.
W as your Physics experience
hands-on?
Yes (go to 8)
No (go to 6)
6.
Are you confident in your ability to
learn Physics?
Yes (go to 9)
No (goto 10)
ilssssg
!!Iilllill
11
H
7.
Do you think you’ve retained som e of
basic knowledge of Physics?
Yes (stop)
No (stop)
8.
Do you tend to work through
assignm ents quickly?
Yes (slop)
No (slop)
9.
Are you generally outgoing or vocal
in group situations?
Yes (go to 8)
No (stop)
Yes (stop)
No (stop)
10. Are you generally outgoing or vocal
in group situations?
FIG U R E 2: Categorization instrument used to determine student group learning
styles in the Fall 1998 semester.
Problems with the Categorization Instrument
When the instrument was administered to the students in the Fall 1998 semester,
several problems were revealed. The most obvious problem seemed related to the
instrument structure itself. Many students failed to pay attention to the instructions telling
them to proceed to a specific, non-consecutive question, often answering all ten questions,
' regardless o f their applicability. For instance, a student with no physics background
answered questions asking about the experiences in the physics class(es) they had taken.
This happened in spite o f repeated explicit verbal instructions telling them to answer only .
the questions directed by their previous answers. Should this particular question tree
structure be used again, a more successful approach might be to simply have students
circle answers along the tree itself. A pictorial description might eliminate the propensity
to answer all the questions.
A second^ more subtle, problem appeared when it was discovered that none o f the
students had classified themselves as passive-uncertain. There are two potential sources
for this discrepancy: either no passive-uncertain students enrolled in the Fall 1998
semester, or the nature o f the questions precluded students from identifying themselves as
such. Considering student descriptions o f their own attitudes, as well as the behavior o f
their fellow group members, it is possible that neither section contained any passiveuncertains. However, we must consider the possibility that the instrument is flawed.
Further examination o f the question tree in Figure I on page 94 and the instrument
in Figure 2 on page 95 reveal two questions that'may have presented problems. Question
3C in Figure I, “Are you confident in your ability to learn Physics? ”, may not be
appropriate in this context. It is reasonable to assume that many students entering a
physics course would only be there if they displayed some confidence in their ability to
learn. If one notes the structure o f the question tree, it can be seen that question 3 C is
along only one o f two paths leading to a cooperative-uncertain categorization. Since no
students who answered question 3B (the other path to a cooperative-uncertain
categorization) answered it in the affirmative, it would appear that this one question not
only eliminated all passive-uncertain categorization, but led to the categorization o f all the
cooperative-uncertains. The main purpose o f this question was to ferret out students’
self-confidence as science learners; it may not have accomplished this purpose. An
improvement to the question might be “Do you think o f yourself as a science person? ”
This wording should be tested, however, as it may make students with no science
background overly prone to not classify themselves as cooperative-uncertains.
The other question which presents some difficulty is question 4B (also 4C) in the
question tree: “Do you tend to be outgoing or vocal in group situations?
While there is
no evidence to show that students felt primarily biased towards one answer, further
consideration leads us to believe the question could be misinterpreted depending on how
the student interpreted the phrase “group situations.” In a close peer group, even an
extremely shy student might be outgoing. M ore appropriate wording should emphasize
the fact that students will probably not be grouped with close friends. Therefore, we feel
the question might best be modified to read, “When meeting people fo r the first time, do
you tend to be outgoing? ”
98
In addition to the two questions mentioned above, it might be appropriate to also
reconsider question 4 ' o f the question tree: “Do you fin d that, on your own, you work
through assignments - quickly? - a t a slower pace? ” In this case, though, the
modification is derived from a greater desire for political correctness. It is possible that
some students will interpret the word “slower” as mildly derogatory, even if it the concept
is accurate in the context o f the question. Therefore, an appropriate rewording o f the
question might be: “A t what pace do you work through assignments? - 1 work quickly. -
I take my time. ’’ This would eliminate the use o f the word “slow.”
A modified version o f the question tree is shown in Figure 3 on page 99. We
anticipate that the changes imposed will improve its accuracy and lead to more consistent
categorization when it is used as an instrument in future classes. -
QUESTION TREE FOR PHYSICS 201 (MODIFIED)
I. H av e you ever
ta k e n P h y sic s b e f o r e ^
2B . D o y o u h av e a n y casual
e x p e rie n c e w ith P h y s ic s?
2 A . D o e s y o u r P h y sic s
e x p e r ie n c e ste m
(e le c tro n ic s o r a s tro n o m y ,
fr o m a h ig h sch o o l
y f o r e x a m p le )
P h y sic s c o u rs e ? .
3 B . I I a s y o u r e x p e rie n c e
3 A . D id y o u h a v e a
p o s itiv e e x p e rie n c e
in v o lv e d h a n d -o n
j n th is P h y s ic s c o u rs e ?
e x p e r im e n ta tio n ? X
SC. Do you think o f
yourself as a
science person^
4B. W henmeetingpeople
fo r the first time, do
tend to be outgoing?
4 A . D o y o u feel y o u ’v e
re ta in e d s o m e o f th e
b a s ic in fo rm a tio n
4C. IfHen meeting people
fo r the first time, do
tend to be outgoing?
p re s e n te d in th is
. P h y s ic s c o u rs e ? X
4 . A t what pace do yox
work through .
assignments?
\
QUfCKLY.
PreviouslyKnowledgeable
AggressiveConfident
CooperativeConfident
CooperativeUncertain
PassiveUncertain.
FIG U R E 3: Modified question tree derived from the discussion on the Appendix. Questions which differ from those in the original
question tree are shown in italics.
44 6T 2241
12/99 30568- 4G ««.6